10-MAY-78 22:13:29-PDT,634;000000000001 Mail from MIT-AI rcvd at 26-SEP-76 0428-PDT Date: 26 SEP 1976 0112-EDT From: KLH at MIT-AI (Ken Harrenstien) Subject: MSGGROUP# 401 Gripe for MSGGROUP distribution To: Stefferud at USC-ISI Message-ID: <4290.[MIT-AI] 09/26/76 01:12:42> Edited-by: Stefferud for MsgGroup What do people think about the inclusion of both FROM: and SENDER: or SENT-BY: in a message, for the common case when both are identical? I think the more elegant algorithm is to only specify SENDER: when it differs from FROM: - otherwise it makes very little sense to ship it along. I've noticed this coming from several places. -Ken ------- 10-MAY-78 22:13:30-PDT,1896;000000000001 Mail from USC-ISIB rcvd at 26-SEP-76 1046-PDT Date: 26 SEP 1976 1043-PDT Sender: COHEN at USC-ISIB Subject: MSGGROUP# 402 Re: MSGGROUP# 401 Gripe for MSGGROUP distribution From: COHEN at USC-ISIB To: STEFFERUD at USC-ISI, MsgGroup at ISI Message-ID: <[USC-ISIB]26-SEP-76 10:43:15-PDT.COHEN> Edited-by: Stefferud In response to your message sent 26 SEP 1976 1014-PDT KEN, RE: the "from" and "sender" fields. I love the separation of "from" and "sender". However, it leaves something to be desired - which I have no solution for: The "sender" is considered as a valid mailbox (really a user), and the "from" is an arbitrary string. This allow my secretary to send a message, which is marked as being "from" me. However, since my name there is an arbitrary string (DANNY COHEN, rather than COHEN) one can not reply to it automatically (like HERMES's REPLY, MSG's ANSWER etc). This results in a communication between the secretaries, rather than between the real correspondants, which is fine most of the time, but not always (for example, I never check my secretary's mailbox, and anything that she gets for me is not accessable when she is away, all 168-40 hours a week). In addition, messages which are important for the record, look nicer if they are addressed to the real addresse, rather than to their secretaries. Mumble... Perhaps the rigt thing to do is to have a "send answer to" field, which is a flag telling if the sender or the from is the right address for answering. All I need is a bit telling if the "from" is a valid mailbox, and if so - I like to send my answer there, else to the sender, with the addition of something like "Attention ". Obviously, this flag can be eliminating and the replier can check if the FROM is really a valid address, by trial and error.... mumble, again. Danny. ------- 10-MAY-78 22:13:30-PDT,953;000000000001 Mail from USC-ISI rcvd at 26-SEP-76 1212-PDT Date: 26 SEP 1976 1203-PDT Sender: STEFFERUD at USC-ISI Subject: MSGGROUP# 403 Idiosyncrasies! Bah! [Mailer:][Mailer:][Mailer:] From: STEFFERUD at USC-ISI To: CAHCOM Cc: Stefferud, MsgGroup Message-ID: <[USC-ISI]26-SEP-76 12:03:38.STEFFERUD> Begin forwarded messages -------------------- Date: 26 SEP 1976 1050-PDT To: STEFFERUD From: Mailer Mail for DEHall at MIT-MULTICS not deliverable because: - Multics idiosyncrasy ------ -------------------- Date: 26 SEP 1976 1050-PDT To: STEFFERUD From: Mailer Mail for Frankston at MIT-MULTICS not deliverable because: - Multics idiosyncrasy ------ -------------------- Date: 26 SEP 1976 1050-PDT To: STEFFERUD From: Mailer Mail for Pogran at MIT-MULTICS not deliverable because: - Multics idiosyncrasy ------ -------------------- End forwarded messages ------- 10-MAY-78 22:13:30-PDT,1526;000000000001 Mail from USC-ISI rcvd at 26-SEP-76 1934-PDT Date: 26 SEP 1976 1923-PDT Sender: STEFFERUD at USC-ISI Subject: MSGGROUP# 404 Re: Error while processing message From: MSGGROUP at USC-ISI To: COMMUNICATION-DAEMON at MIT-DMS, MSGGRP at MIT-DMS, To: Vezza at MIT-DMS Cc: MsgGroup, Stefferud, Broos Message-ID: <[USC-ISI]26-SEP-76 19:23:36.STEFFERUD> In-Reply-To: <[MIT-DMS].40682> I do not understand any cause for this, nor am I sure to whom this was addressed. Your move next. Enjoy, Stef -------------------- Mail from MIT-DMS rcvd at 26-SEP-76 1607-PDT DATE: 26 SEP 1976 1855-EDT FROM: COMMUNICATION-DAEMON at MIT-DMS SUBJECT: Error while processing message KEYWORDS: ERROR, BUG, PROCESS MESSAGE-ID: <[MIT-DMS].40682> An error occurred while processing a message which may be of concern to you. The message id was <[USC-ISI]26-SEP-76 10:14:50.STEFFERUD> . The message was from MSGGROUP at USC-ISI. The subject was [KLH@MIT-AI (Ken Harrenstien): MSGGROUP# 401 Gripe for MSGGROUP distribution]. The PARSING process failed because a MUDDLE error occurred during the processing. This indicates a possible bug, and has been reported to the system maintainer(s). The problem has also been reported to the message system maintainer(s). If you do not understand the reasons for the problem, send a message to the COMMUNICATION-SYSTEM-MAINTAINER for help. Please include the id of the failing message. -------------------- ------- 10-MAY-78 22:13:30-PDT,1061;000000000001 Mail from MIT-AI rcvd at 26-SEP-76 2115-PDT Date: 27 SEP 1976 0015-EDT From: Richard M. Stallman (RMS @ MIT-AI) Subject: MSGGROUP# 405 RE: MSGGROUP# 401 Gripe for MSGGROUP distribution To: Stefferud at USC-ISI, MSGGROUP at USC-ISI Message-ID: <4591.[MIT-AI] 09/27/76 00:15:40> Edited-by: Stefferud I was surprised to hear that other systems send replies automatically to the SENDER of a message. We send them to the FROM, since our normal usage of SENDER and FROM is that the FROM is the user who really sent the message, while the SENDER is the user whose terminal he was using at the time. The only interest of the SENDER field is that it says who to make a com-link with if you want to communicate immediately with the other user. Thus, it seems that the meanings of FROM and SENDER are understood differently at different places, and perhaps more alternatives are needed than those two. In any case, the various forms of author should all have defaults and only those containing significant information should be sent. ------- 10-MAY-78 22:13:30-PDT,2267;000000000001 Mail from OFFICE-1 rcvd at 28-SEP-76 1556-PDT Date: 28 SEP 1976 1543-PDT From: HOUGH at OFFICE-1 Subject: MSGGROUP# 406 Second Computer Inquiry To: [ISI]Mailing.List: cc: gedwards, bedford, napke cc: rulifson at PARC-MAXC, Taylor at PARC-MAXC, cc: Sutherland at PARC-MAXC < HOUGH, FCC.NLS;1, >, 28-SEP-76 15:21 RWH ;;;; The Second Computer Inquiry is now underway at the Federal Communications Commission. If you wish to enter comments, you must do so by October 12, 1976. In the First Computer Inquiry, which ended in 1971, the FCC ordered that all "message switching" services apply for common carrier status. While the FCC's definition of message switching was not crystal clear, it seemed to embrace computer mail box services such as MSG and Hermes and computer conferencing services such as FORUM. In the proposed rules of the new inquiry, it appears -- and this is only my initial interpretation of the rules -- that the FCC has not only removed computer mail and conferencing services from common carrier status but has even forbidden common carriers such as Western Union, AT&T and some specialized common carriers from offering computer mail and conferencing services, except through an unregulated subsidiary. Specifically, common carriers would be prohibited from offering "word processing" services, which would seem to include composition, editing, filing and retrieving, except through an unregulated subsidiary. Only network transmission, for example ARPANET's role in current systems, would remain the bailiwick of common carriers. Since AT&T is prohibited, under the 1956 consent decree, from entering nonregulated markets, it would seem to be prohibited from offering computer message services. I would appreciate your comments on my interpretation of the proposed rules. If there is sufficient interest in this topic, we should start a separate working group. By the way, my draft paper, The Outlook for Computer Message Services: a Preliminary Assessment, which many of you have, has an appendix on the First Computer Inquiry. Ra3y Panko Stanford Research Institute 333 Ravenswood Ave. Menlo Park, CA, 94025 (415) 326-6200, ext. 4213 ------- 10-MAY-78 22:13:31-PDT,335;000000000001 Mail from USC-ISI rcvd at 28-SEP-76 2305-PDT Date: 28 SEP 1976 2252-PDT Sender: STEFFERUD at USC-ISI Subject: MSGGROUP# 407 $Please remove DINGMAN@OFFICE-1 from list$ From: MSGGROUP at USC-ISI To: [ISI]Mailing.List: Cc: Dingman at OFFICE-1 Message-ID: <[USC-ISI]28-SEP-76 22:52:10.STEFFERUD> Stef ------- 10-MAY-78 22:13:31-PDT,2098;000000000001 Mail from OFFICE-1 rcvd at 29-SEP-76 0936-PDT Date: 29 SEP 1976 0936-PDT From: SUTKOWSKI at OFFICE-1 Subject: MSGGROUP# 408 text editor group ideas To: dick at ILL-NTS, gls at MIT-AI, rms at MIT-AI, To: sunshine at RAND-UNIX, weiner at RAND-UNIX, To: gammill at RAND-UNIX, dcrocker at RAND-UNIX, tugender at ISIB, To: oestreicher at ISIB, uk at ISI, map at MIT-MULTICS, potter, To: bailey, farber at ISI, stefferud at ISI, msggroup at ISI, To: geoff at SRI, hathaway at AMES-67, hornish at BBN, To: jf at CCA, kahler at SUMEX-AIM, dornbush, aloha at ISID, To: westdiv at ISI, lipner at ISIA, robinson, stone, To: tlr at SU-AI, stillman at RUTGERS-10, pipes at I4-TENEX, To: mccall at SRI, plummer at BBNA, sutton at BBNA To all Text-Editor Group Members: Before compling the names and setting up the mailing list, i had talked with E. Stefferud, coordinator of the MSGGROUP mailing list at isi, and he made a few suggustions that i believe could be incorporated into this group. They are as follows: 1) Txedgrp.memb and Txed*nnn, could be set up as the group-name. mailing list and "acession number" code respectively (i.e. each message is numbered after having first been presented to the group) 2) Identifying info should be placed in the TO, FROM, SUBJECT fields, so as to conform to the mailing systems of various sites. 3) Set up a directory for the editor group (here at office-1) similar to the MSGGROUP directory. 4) A possibility of the text-editor group becoming a subgroup of msggroup. Comments anyone? Steve P. S. here is the membership list as it stands at the present time: AMES-67: Hathaway BBN: Hornish BBNA: Plummer, Sutton CCA: JF I4-TENEX: PIPES ILL-NTS: DICK ISI: Uk, Farber,Stefferud,MSGGROUP,Westdiv ISIA: LIPNER ISIB: Tugender, Oestreicher ISID: Aloha MIT-AI: GLS, RMS MIT-MULTICS: MAP OFFICE-1: Potter, Bailey, Dornbush, Robinson, Stone, Sutkowski RAND-UNIX: Sunshine, Weiner, Gammill, Dcrocker Rutgers-10: Stillman SRI-AI: Geoff, Mccall SU-AI: TLR SUMEX-AIM: Kahler ------- 10-MAY-78 22:13:31-PDT,886;000000000001 Mail from OFFICE-1 rcvd at 27-SEP-76 0952-PDT Date: 27 SEP 1976 0952-PDT From: SUTKOWSKI at OFFICE-1 Subject: MSGGROUP# 409 text-editor group To: stefferud at ISI This is Steve Sutkowski, presently at Eglin afb. As many of you may be aware, I am starting up a text-editing group comprising individuals of the following types: 1) Contributors: those having specific ideas or innovations in the field of text editing, either in regards to their own system or others, or those people who note certain faults or drawbacks to editing which they would like to point out. 2) Observers: those interested in whats going on in the field, who perhaps have limited knowledge or experience but wish to find out more. If by any chance you may be interested in this, please let me know. You may contact me via sutkowski at office-1. thanks. Steve Sutkowski ------- 10-MAY-78 22:13:31-PDT,1305;000000000001 Mail from USC-ISIC rcvd at 29-SEP-76 1457-PDT Date: 29 SEP 1976 1450-PDT From: POSTEL at USC-ISIC Subject: MSGGROUP# 410 Re: from, sender, and standards. To: stefferud at ISIA cc: postel REFERENCE: MSGGROUP# 401, MSGGROUP# 402, MSGGROUP# 405 [Stef: Please forward to msg-group.] The SRI-ARC sendmail system has used the terms Author and Clerk instead of From and Sender. Author is the person who had the ideas and Clerk is the person who pushed the buttons. The official protocol is that defined in RFC 680. RFC 680 indicates that From is the person who "wished this message to be sent" while Sender is the person who "sends the message". In sendmail's terms Sender = Clerk and From = Author. Let me repeat that RFC 680 is the existing standard for header formats. It is nice that many mail reading programs will accept mail that does not comform to the standard but that does not justify mail sending programs' violation of the standard. If the standard is inadaquate for applications we desire to pursue lets hear specific proposals for modification of the standard. However if substantial modifications are necessary then i am much inclined in the direction of Jack Haverty et.al.'s work on a different approach to network mail. --jon. ------- 10-MAY-78 22:13:31-PDT,1617;000000000001 Mail from OFFICE-1 rcvd at 30-SEP-76 0915-PDT Date: 30 SEP 1976 0915-PDT From: HOUGH at OFFICE-1 Subject: MSGGROUP# 411 Moist, Damp, From, To, Author, and Clerk To: [ISI]Mailing.List: REFERENCE: MSGGROUP# 410 One time, when I was writing a report, I used the word "moist." After a few minutes, I decided that the word "damp" would be better. During the next day, I shifted back and forth at least a dozen times. In the end, I just decided that moist and damp were too close for even an author to discern, much less a reader. I think that "sender" and "from" have the same problem. If you tell me which is which, I could easily make up some mnemonic or image to help me get the difference straight. After awhile, the distinction would seem natural. Yet for the novice, I think that the words "from" and "sender" have connotations too close to make them useful. I like the word clerk, because it has a nice connotation of subordinate. Secretary would be nice too, but secretaries send messages, and because furture mail systems will probably handle forms and data as well as correspondence, clerk might be better. The term author seems to have a nice ring, because an author is somebody who puts his or her name on a finished document. But again, it may be that author has a clear connoation only after you think about it. Just a thought..does anybody know how the sender/from distinction is traditionally handled in military arenas and in services like Telex and TWX, where composition and sending are separated from authorship? Ra3y Panko ------- 10-MAY-78 22:13:31-PDT,608;000000000001 Mail from USC-ISI rcvd at 30-SEP-76 1011-PDT Date: 30 SEP 1976 1002-PDT Sender: ELLIS at USC-ISI Subject: MSGGROUP# 412 RE: The "FROM" discussion From: TOM ELLIS To: [ISI]Mailing.List: Message-ID: <[USC-ISI]30-SEP-76 10:02:46-PDT.ELLIS> Special-Handling: ATTN: PANKO@OFFICE-1 In military circles, the term "FROM" designates the authority behind the message. It often, if not always, carries the top title of the organization (not a personal name) and implies that the message has at least been reviewed and formally released by the person holding that office. ------- 10-MAY-78 22:13:31-PDT,3958;000000000001 Mail from MIT-DMS rcvd at 30-SEP-76 1016-PDT DATE: 30 SEP 1976 1311-EDT FROM: JFH at MIT-DMS SENDER: JFH at MIT-DMS SUBJECT: MSGGROUP# 413 RE your message to MSGGROUP about from and sender... ACTION-TO: HOUGH at OFFICE-1, HEADER-GROUP at MIT-MC CC: STEFFERUD at USC-ISI, FARBER at USC-ISI MESSAGE-ID: <[MIT-DMS].40984> Edited-by: Stefferud Reference: MSGGROUP# 411 Note: To STEFFERUD@USC-ISI, FARBER@USC-ISI, HOUGH@OFFICE-1: Please forward this to people who might be interested. (I sure wish I could send to things like [USC-ISI]mail-list... Editors Note: Anyone can send to the [ISI] Mailing.List by simply sending one copy of the message to MSGGROUP@ISI, with another copy to STEFFERUD@ISI to alert me to foreward it. I will insert the MSGGROUP# and then forward it. S/ Message: 1/ I believe CLERK was proposed at one time in the distant past, and rejected in favor of SENDER because it was thought that CLERK would be somewhat demeaning to the person involved. 2/ There are lots of other possible people (not necessarily all distinct) associated with a message. For example, RELEASED-BY -- the person who said it is ok to send AUTHORIZED-BY -- the person who approves of the 'content' SENT-BY -- what SENDER is now REPLY-TO -- who should get the answers REQUESTED-BY -- who asked that the message be composed, e.g. a manager may ask a staff member to compose some information into a message AUTHORED-BY -- the person who actually wrote the stuff COLLABORATORS -- people who assisted the AUTHORED-BY CO-AUTHORS -- others who assume equal responsibility with the AUTHORED-BY etc. etc. etc. In all of these cases, it is reasonable (except for SENT-BY) to allow any number of names, etc. 3/ The concept of a 'person' is inadequate. Often a 'position' is also necessary. For example, someone can act as himself in a personal sense, as holder of some office within an organization, etc. This can significantly change the meaning of anything such a pseudo-schizophrenic may say. In the military, the problem is handled by requiring that all messages be from 'the commander' in some fashion -- e.g. this message would be from 'MIT-DMS', not 'JFH@MIT-DMS' 4/ The common occurrence in current mail where a person sends a message using another account complicates things further. For example, the message which Panko sent was signed by him, but the header alleged it to be from HOUGH@OFFICE-1. ------ A comment: The problem is much more complex than most people think. It is not clear that a 'standard' can be created which simultaneously satisfies the needs of all users on the various sites. Even if it can, it may be prohibitively expensive, if the vast majority of messages fit some simple form, e.g. a short note from a single real live person to another single person, with no replies, references, etc. involved. Systems which are forced to consider all the possibilities all the time are necessarily more complex, difficult to program, maintain, debug, etc. A proposal: Has anyone ever considered that there is no inherent need for a single MAIL SYSTEM, or even a STANDARD HEADER. Possibly it would be more reasonable to consider having several separate mail paths, each suited to a different class of message/user. The real world analog has the US Postal Service, UPS, Air freight, Western Union, Carrier Pigeon, Radio, and so on. It might be worth thinking about. A lot of problems associated with changeovers in THE STANDARD could also be avoided if new systems were phased in and the old allowed to die through disuse over time. I'm not sure what ramifications this would have at the various sites -- presumably each service would require a separate 'port' (socket to ICP to, or command to direct a message to various handlers for different systems). Any comments? Jack Haverty (JFH@MIT-DMS) ------- 10-MAY-78 22:13:32-PDT,442;000000000001 Mail from OFFICE-1 rcvd at 30-SEP-76 1033-PDT Date: 30 SEP 1976 1033-PDT From: SUTKOWSKI at OFFICE-1 Subject: MSGGROUP# 414 rsx-11m difficulties: request for info To: msggroup at ISI I'm wondering if anyone in the group has worked with the line text editor of the rsx-11m operating system, and if so could connect me in regards to it, since i have some specific questions i'd like to ask. thanks. Steve Sutkowski ------- 10-MAY-78 22:13:32-PDT,2159;000000000001 Mail from OFFICE-1 rcvd at 30-SEP-76 1223-PDT Date: 30 SEP 1976 1147-PDT Sender: VONGEHREN at OFFICE-1 Subject: MSGGROUP# 415 RE: MSGGROUP "FROM" DISCUSSION From: E. VONGEHREN To: STEFFERUD at ISI Message-ID: <[OFFICE-1]30-SEP-76 11:47:43.VONGEHREN> FRANKLY I FIND IT MUCH MORE INTERESTING TO NOTE WHAT TOM ELLIS DID WITH "HIS" FROM AND SENDER FIELDS (IN THIS MEDIA) THAN TO PONDER WHAT THE MILITARY DO WHEN THEY SEND OUT FORMAL CORRESPONDENCE. I THINK THAT WE ARE TOO EASILY LEAD BY WHAT SPECIALIZED GROUPS HAVE DONE IN ANOTHER MEDIUM AND INFER THAT MIGHT PROBABLY BE DONE IN GENERAL IN THIS MEDIUM. BEFORE KNOCKING THE WORDS ABOUT LET'S DROP BACK AND CONSIDER THE ALTERNATIVE SITUATIONS WHICH ARE REASONABLE TO ADDRESS. 1) I COMPOSE MY MESSAGE AND I SEND IT. 2) I COMPOSE MY MESSAGE AND MY SECRETARY SENDS IT. 3) I COMPOSE A MESSAGE FOR MY BOSS (OR ORGANIZATION), MY BOSS APPROVES (SIGNS) IT, MY SECRETARY SENDS IT. 4) ETC. WHAT THIS IS ALL LEADING UP TO IS THAT THERE ARE A VARIETY OF POSSIBLE PEOPLE (AND NON-PEOPLE SUCH AS ORGANIZATIONS) INVOLVED IN THE PREPARATION OF CORRESPONDENCE. YOU MAY WANT TO IDENTIFY THEM ALL TO THE READER, AND THEN YOU MAY NOT; YOU MAY ONLY WANT TO MAINTAIN THEIR IDENTITY FOR YOUR PERSONAL RECORDS. QUESTION: DO WE AS A DISCUSSION GROUP REALLY WANT TO DISCUSS AN ALL GENERALIZED "CORRESPONDENCE" SYSTEM OR ARE WE NIT PICKING WORDS THAT WE'VE BEEN TOLERANT OF IN THE PAST. CONSEQUENCE: IF THE ANSWER TO THE QUESTION IS THAT WE WANT TO DISCUSS THE DESIGN OF AN ALL GENERALIZED CORRESPONDENCE SYSTEM THEN I THINK WE SHOULD PUT OUR EFFORTS IN TO RESOLVING SUCH ISSUES AS: - "WHAT IS THE BEST WAY TO STRUCTURE MY MAIL PREPERATION AND FILING SYSTEM SO THAT I CAN RETAIN THESE SPECIAL NOTATIONS IN MY FILED COPY THAT DID NOT GET SENT OUT WITH THE ORIGINAL COPY, AND CAN I FORWARD A COPY OF WHAT I HAVE ON FILE WITHOUT SENDING THESE ADDITIONAL NOTATIONS?" PEACE- ED ------- 10-MAY-78 22:13:32-PDT,2961;000000000001 Mail from USC-ISI rcvd at 1-OCT-76 1135-PDT Date: 1 OCT 1976 1110-PDT Sender: CAHCOM at USC-ISI Subject: MSGGROUP# 416 The first of a series of messages Subject: outlining the activities of the CAHCOM committee From: CAHCOM at USC-ISI To: [ISI]Mailing.List: Message-ID: <[USC-ISI] 1-OCT-76 11:10:00.CAHCOM> The purpose of this series of messages is to better acquaint the Message Group membership with the current activities of the ARPA IPTO CAHCOM (Computer Aided Human Communication) committee. We hope that some of these activities will excite you to contribute your help and insights to our effort. First a bit on the organization of the CAHCOM committee. It has a steering committee composed of Dave Farber of UC Irvine (Chairman), Bob Anderson of RAND, John S. Brown of BBN, Danny Cohen of ISI , Ken Pogran of MIT-LCS, Al Vezza of MIT-DMS , John Vittal of BBN and Steve Walker of ARPA IPTO. The steering committee has created a number of working groups (sub-committees) which have been charged to operate in specific areas in message technology and the IT . One of these formed on August 3 1976 is charged with examining and recommending standards in the area of message headers . It is co-chaired by Ken Pogran and John Vittal and has as members Austin Henderson of BBN and Dave Crocker of RAND-USC. The charter of their sub-committee is attached to this message. I would encourage your help in their task and look forward to profitable interactions and collaboration between you and CAHCOM. Dave Farber The ARPA IPTO Committee on Computer-Aided Human Communication (CAHCOM) believes it important that unified development of message service facilities in the ARPA Network community be encouraged, especially from the protocol/syntactic points of view, and that these protocols become official ARPA Network protocols. This subcommittee of CAHCOM is charged with the task of developing a new standard for the format of ARPA Network message headers which will adequately meet the needs of the various message service subsystems on the Network today. The subcommittee is addressing two primary issues: 1. A re-specification of message header formats so as to incorporate the facilities which people have found missing in the present standard, published as RFC-680. 2. Turning this new specification into an official ARPA Network protocol standard. This will be done only after the proposal developed by the subcommittee has been widely circulated, reviewed by message service implementors and other interested members of the ARPA Network community, and revised to reflect the input obtained from these sources. Also, the subcommittee will endeavor to obtain the agreement of message service implementors to implement and adhere to the new protocol. ------- 10-MAY-78 22:13:32-PDT,5645;000000000001 Mail from USC-ISI rcvd at 3-OCT-76 0139-PDT Date: 3 OCT 1976 0119-PDT Sender: STEFFERUD at USC-ISI Subject: MSGGROUP# 417 Sept '76 Transactions From: MSGGROUP at USC-ISI To: [ISI]Mailing.List: Message-ID: <[USC-ISI] 3-OCT-76 01:19:40.STEFFERUD> Special-Handling: Requests for specific msgs will receive Special-Handling: forwarded copies. Stef -- Messages from file: [USC-ISI]TRANSACTIONS.MSG;6 -- SUNDAY, OCTOBER 3, 1976 01:05:03-PDT -- 416 1 OCT CAHCOM at USC-ISI MSGGROUP# 416 The first of a series of messages (2961 chrs) 415 30 SEP E. VONGEHREN MSGGROUP# 415 RE: MSGGROUP "FROM" DISCUSSION (2159 chrs) 414 30 SEP SUTKOWSKI at OFFICE-1 MSGGROUP# 414 rsx-11m difficulties : request for info (442 chrs) 413 30 SEP JFH at MIT-DMS MSGGROUP# 413 RE your message to MSGGROUP about from and sender... (3958 chrs) 412 30 SEP TOM ELLIS The message which you sent to AV at MIT-DMS has not yet been delivered. ------- 10-MAY-78 22:13:32-PDT,2294;000000000001 Mail from OFFICE-1 rcvd at 7-OCT-76 1711-PDT Date: 7 OCT 1976 1711-PDT From: HOUGH at OFFICE-1 Subject: MSGGROUP# 420 The Second Computer Inquiry - Some Definitions To: [ISI]Mailing.List: If there is much more traffic about the Second Computter Inquiry, we should start a new working group. Essentially, the rules proposed in the new inquiry would require that common carriers offerd data processing services (as defined in the inquiry) only through unregulated subsidiaries. But - Catch 22 -- AT&T is precluded from offering nonregulated services, under its 1956 Consent Decree entered into with the Justice Department. Specifically, any services involving the following would be considered to be data processing: ARITHMETIC PROCESSING - general commercial accounting, payroll, inventory control, banking and point-of-sale-processing, financial and econometric modeling, scientific calculations, etc. WORD PROCESSING - a rapidly developing application resulting from advances in mass memory technology and word processing software, such as, interactive information retrieval systems, management information systems, text editing, translation, type-setting, etc. PROCESS CONTROL - the increased reliability and availability of computers is leading to an expansion of applications where a computer is used to monitor and control some process which is occurring continu- ously, such as nuclear-power generating stations, an electric power distirubtion grid, an automatic machine tool, or a fire detection and control system. On the other hand, common carriers will bbe allowed to offer bare- bones bit transmission, such as: NETWORK CONTROL AND ROUTING - including message and circuit switching , speed and code coversion, pulse format conversion, error detection and correction, analog to digital and digital to analog conversion, signal processing, and time division multiplexing; INPUT/OUTPUT PROCESSING - using computer capability in a carrier network or facility to make different computers and terminals compatible with one another. Under these conditions, I (Ra3y Panko) believe that computer message services that involve any form of composition, filing or reading tools fall under data processing. ------- 10-MAY-78 22:13:33-PDT,1111;000000000001 Mail from USC-ISI rcvd at 8-OCT-76 1245-PDT Date: 8 OCT 1976 1228-PDT Sender: DCROCKER at USC-ISI Subject: MSGGROUP# 421 Re: Comment on FCC Second Computer Inquiry From: DCROCKER at USC-ISI To: MSGGROUP, To: Pogran at MULTICS Cc: [ISI]Mailing.List:, Cc: [isi]hic.folk: Message-ID: <[USC-ISI] 8-OCT-76 12:28:56.DCROCKER> In-Reply-To: <[USC-ISI] 7-OCT-76 14:55:33.STEFFERUD> REFERENCE: MSGGROUP# 418 It should be noted, for those not yet aware, that AT&T is attempting an end-run around the FCC. The FCC has not been terribly friendly to the Big A (cf. the Carterfone decision, allowing non AT&T units attached to the phone system); so AT&T has introduced the Consumer Communications Reform Act of 1976 into Congress. The likely result of its passage has been compared to the British Post Office, which is a COMPLETE monopoly. AT&T, of course, does not agree that it seeks control to such an extreme, but others disagree. It is fairly clear that if the CCRA'76 is passed, many companies -- Telenet included -- would go under. Dave. ------- 10-MAY-78 22:13:33-PDT,1213;000000000001 Mail from OFFICE-1 rcvd at 9-OCT-76 1406-PDT Date: 9 OCT 1976 1406-PDT From: PANKO at OFFICE-1 Subject: MSGGROUP# 422 Meeting on the FCC Computer Inquiry and a comment on the "Bell Bill" To: [ISI]Mailing.List: On November 8th and 9th, there will be a conference on computers and communication in Washington, D.C., to help the FCC understand aspects of computers that will affect the current Computer Inquiry. If you are interested, I can supply somewhat more information. I believe that the conference is sponsored by IFIPS. Don Dunn of Stanford and Vint Cerf (now) of ARPA are the honchos. Since I am consulting for AT&T on the Inquiry, I am somewhat retrained from making comments about the proceedings that could be interpreted as advocacy. But I will be happy to discuss the proceedings informally, if any of you want to attend the planning conference. I don't believe the "Bell Bill"" now being considered in Congress is likely to pass, and I haven't seen any versions that are completely anticompetitive. It is true, however, that passage f the bell bill would have tremendous implications of the data processing community. That is my informal opinion. ------- 10-MAY-78 22:13:33-PDT,1237;000000000001 Mail from BBN-SPEECH11 rcvd at 13-OCT-76 0914-PDT Date: 13 Oct 1976 1201-EDT Sender: HENDERSON at BBN-TENEXF Subject: MSGGROUP# 423 Change of address from BBNA to BBNF From: HENDERSON at BBN-TENEXF To: [ISI]Mailing.List:, To: [BBNF]HDISCUSS.MAILINGLIST:, To: AML.Mailing-List;4:, To: [ISI]DELIMITER.LIST:, To: [BBNF]HERMES.1ST-DIST-LIST: Message-ID: <[BBN-TENEXF]13-Oct-76 12:01:29.HENDERSON> To whom it may concern: Those of us working in Project Hermes at BBN have changed our network addresses. We are now on "BBNF". Please change all your address lists (and heads) to reflect this fact. We would appreciate it if you would do this fairly quickly, since the mail forwarding system on BBNA (our old host) is not working at the moment, and will take a few days to get working. Thus mail addressed to BBNA will go undelivered. Thank you for your attention to this matter. Austin Henderson John Vittal Charlotte Mooers Ted Myer Doug Dodds Frank Ulmer Jim Miller PS: BBNF has host number 62(8) = 50(10). At one time that host number was assigned to "BBN-SPEECH-11", so you might find it in your tables under that name. ------- 10-MAY-78 22:13:33-PDT,425;000000000001 Mail from USC-ISI rcvd at 15-OCT-76 0552-PDT Date: 15 OCT 1976 0540-PDT Sender: SLES at USC-ISI Subject: MSGGROUP# 424 Lipner now = Sles at ISIA From: Sles To: [ISI]Mailing.List: Message-ID: <[USC-ISI]15-OCT-76 05:40:41.SLES> Please update the MsgGroup mailing list--Steve Slesinger and Steve Lipner still share an account, but the name of the directory is now SLES, not LIPNER. Thanks. ------- 10-MAY-78 22:13:33-PDT,1625;000000000001 Mail from USC-ISI rcvd at 18-OCT-76 2007-PDT Date: 18 OCT 1976 2003-PDT Sender: STEFFERUD at USC-ISI Subject: MSGGROUP# 425 Changes in [ISI]MAILING.LIST;93 From: MSGGROUP at USC-ISI To: [ISI]Mailing.List: Message-ID: <[USC-ISI]18-OCT-76 20:03:07.STEFFERUD> It seems that this is address changing time, so here is a wholesale batch. I am sending the whole list so you can convert it to your uses, or file it. Note the changes are all at the following HOSTs: ISI, RAND-UNIX, RAND-ISD (gone), RAND-RCC, UCLA-SECURITY (new), and BBNA has become BBNF (new). Lipner@ISI is now Sles@ISI, Anderson@RAND-RCC is now anderson@RAND-UNIX, BRD@RAND-RCC is now BRD@RAND-UNIX, Lauren@RAND-UNIX is now Laren@UCLA-SECURITY. The new list follows: [ISI]Mailing.List: @BBN, DBrown, Gilbert, Heitmeyer, Hornish, Mathison, Perlingiero, Walker, @BBNB, Feinler, SMartin, @BBNF, Burchfiel, Henderson, Mooers, Myer, Vittal, @CCA, Tom, @CMUA, Barnes, Karlton, Newcomer, Wactlar, @ISI, MSGGROUP, PBaran, Broos, DCrocker, Farber, Kirstein, Mclindon, Sles, Spivey, Stefferud, Tasker, Walker, @ISIB, Cohen, Ellis, Steve, Stotz, Wulf, @ISIC, Postel, White, @ISID, Carlisle, @MIT-AI, KLH, RMS, @MIT-DMS, MSGGRP, Vezza, @MIT-MULTICS, DEHall, Frankston, Pogran, @MIT-ML, CBF, @OFFICE-1, Dames, Engelbart, Grobstein, Hough, Stone, Sutkowski, Taylor, Uhlig, vonGehren, @PARC-MAXC, McDaniel, @RAND-UNIX, BRD, gaines, mlw, anderson, wec, @RUTGERS-10, Robertson, @SRI-AI, Geoff, @SUMEX-AIM, Kahler, NSmith, @UCLA-ATS, Mark, @UCLA-SECURITY, Lauren, ------- 10-MAY-78 22:13:33-PDT,1255;000000000001 Mail from USC-ISI rcvd at 20-OCT-76 0953-PDT Date: 20 OCT 1976 0929-PDT Sender: FARBER at USC-ISI Subject: MSGGROUP# 426 Am example of the failure of systems maintainance From: FARBER at USC-ISI To: [ISI]Mailing.List: Message-ID: <[USC-ISI]20-OCT-76 09:29:19.FARBER> I think what has happened in TENEX is a excellent example of the fact that people who maintain operating systems quite often do not think of the impact of what they do on users. Recently TENEX maintainers decided to only allow 4 generations of a given file. Surplus generations are deleted. Due to a problem at RAND with their new TIP mail has not been being delivered to them. I have been activly corresponding with Bob Andersoon there and lo it seems that more than 4 messages got queued and thus there is mail to Bob that he will NEVER get and if I did not realize what was going on like certain ARPA executives might not , I would get very angry at BOb when he did not answer my messages which he of course would never have gotten. The running of a utility is difficult. One does not do things that sound good without carefully thinking and talkinbg with major user syste,m implementors. Dave Farber ------- 10-MAY-78 22:13:33-PDT,315;000000000001 Mail from USC-ISI rcvd at 20-OCT-76 1553-PDT Date: 20 OCT 1976 1536-PDT Sender: STEFFERUD at USC-ISI Subject: MSGGROUP# 427 $Please swap Hough@Office-1 for Panko@Office-1$ From: MSGGROUP at USC-ISI To: [ISI]Mailing.List: Message-ID: <[USC-ISI]20-OCT-76 15:36:18.STEFFERUD> Stef ------- 10-MAY-78 22:13:34-PDT,1527;000000000001 Mail from OFFICE-1 rcvd at 28-OCT-76 1910-PDT Date: 28 OCT 1976 1442-PDT From: PANKO at OFFICE-1 Subject: MSGGROUP# 428 Murkier and murkier at the FCC To: [ISI]Mailing.List: cc: martin at USC-ECL, day at OFFICE-1, gedwards at OFFICE-1, cc: napke at OFFICE-1, leduc at OFFICE-1, panko at OFFICE-1, cc: bedford at OFFICE-1 Let me see if I can explain it. Under the First Computer Inquiry, the FCC decreed in 1971 that all message-switching would have to be done by common carriers. Although computer mail and computer teleconferencing were not specifically mentioned in the inquiry, private discussions with the FCC made it clear that conferencing, mail boxing, etc. would fall under the "gotta be offered by common carrier" heading. This year, in its notice of proposed rule making, the FCC PROSCRIBED common carriers from direcely offering "word processing" services, including text editing, information retrieval and other things that all mail systems anywhere near the state of the art always provide. Oh, some of us said, the FCC has removed computer mail from common carrier status. Suspecting that all was not really that clear, I had one of my compatriots contact (informally) the fellow in Common Carrier Bureau who knows most about mail systems. No, he said, computer mail HAS TO BE OFFERED BY COMMON CARRIERS< EVEN UNDER THE NEW RULING (which seems to proscribe them from offering the service). Does anybody else here see a paradox? Ra3y Panko ------- 10-MAY-78 22:13:34-PDT,1016;000000000001 Mail from USC-ISI rcvd at 30-OCT-76 1322-PDT Date: 30 OCT 1976 1314-PDT Sender: STEFFERUD at USC-ISI Subject: MSGGROUP# 429 Add RCT@CCA, Replace Carlisle@ISID with Carlisle@USC-ECL Subject: [TOM at CCA: ADDITION TO MSGGROUP DISTRIBUTION LIST] From: MSGGROUP at USC-ISI To: [ISI]Mailing.List: Message-ID: <[USC-ISI]30-OCT-76 13:14:15.STEFFERUD> Carlisle asked that we send his mail to Carlisle@USC-ECL, and RCT@CCA has been added to the list. Begin forwarded message -------------------- Mail from CCA-TENEX rcvd at 30-OCT-76 0810-PDT Date: 30 OCT 1976 1115-EDT From: TOM at CCA Subject: ADDITION TO MSGGROUP DISTRIBUTION LIST To: STEFFERUD at ISI cc: RCT, TOM STEFF: WOULD IT BE POSSIBLE FOR RUSS TREADWELL AT COMPUTER CORP OF AMERICA, RCT@CCA TO BE ADDED TO THE DISTIRIBUTION LIST FOR MSGGROUP BUSINESS? RUSS WOULD VERY LIKE TO BE KEPT INFORMED. THANKS. TOM MARILL ------- -------------------- End forwarded message ------- 10-MAY-78 22:13:34-PDT,309;000000000001 Mail from USC-ISI rcvd at 1-NOV-76 1101-PST Date: 1 NOV 1976 1046-PST Sender: STEFFERUD at USC-ISI Subject: MSGGROUP# 430 $Change Feinler@BBNB to be Feinler@OFFICE-1$ From: MSGGROUP at USC-ISI To: [ISI]Mailing.List: Message-ID: <[USC-ISI] 1-NOV-76 10:46:18.STEFFERUD> Stef ------- 10-MAY-78 22:13:34-PDT,1049;000000000001 Mail from OFFICE-1 rcvd at 3-NOV-76 0927-PST Date: 3 NOV 1976 0921-PST From: PANKO at OFFICE-1 Subject: MSGGROUP# 431 Request for Information on Computer Message Services To: [ISI]Mailing.List: I am redoing my paper on the outlook for computer message services. I would be extremely grateful if any of you could provide any information on the costs of existing or planned computer message services. I am especially interested on the cost for an average message and the average length of a message. I am also lookin for information on labor costs associated with computer mail. I would also appreciate any references to computer message services I might not be familiar with and any suggestions for policy issues that might affect the emergence of computer mail. It will take me about a month to redo the paper. When the change is complete, I will send around a notice. Those of you who read the March draft will be pleased to learn that this paper will be much shorter. Ra3y Panko at Office-1 ------- 10-MAY-78 22:13:34-PDT,1994;000000000001 Mail from OFFICE-1 rcvd at 5-NOV-76 1333-PST Date: 5 NOV 1976 1330-PST From: PANKO at OFFICE-1 Subject: MSGGROUP# 432 A chat with the FCC on computer mail To: [ISI]Mailing.List: cc: day at OFFICE-1, leduc at OFFICE-1, gedwards at OFFICE-1, cc: bedford at OFFICE-1, rulifson at PARC-MAXC, cc: sutherland at PARC-MAXC, taylor at PARC-MAXC, cc: metcalfe at PARC-MAXC, carlisle at USC-ECL, cc: martin at USC-ECL I had a chat this morning with Ray Dujack of the FCC, on the legality of computer mail. As most of you know, the FCC is now well into the Second Computer Inquiry, which may profoundly affect computer mail. I want to pass on my notes, but please understand that these are only my impressions after talking with Ray, not official FCC policies and perhaps not even an accurate portayal of what Ray actually said. Ray basically concludes that computer mail systems offered for hire must be regulated as common carrier services. Look out, HERMES. At the same time, the FCC isn't going out to enforce its rules in this matter--unless the industry gets to be a substantial size. My original reading of the Commission's current proposed rules is that by forbidding common carriers to offer editing and information retireval the FCC was prohibiting them from computer mail. But although the language is cloudy, it appears that the FCC will still regulate on the basis of the real thrust of the service. It is by no means clear where the FCC will draw the dividing line between permissible editing in a mail box service and impermissively powerful editing. It does seem to me, however, that thhe basic trend for computer systems to serve many DP, documentation and communication needs will be frustrated by the FCC forcing one company to offer only communication services on a common carrier basis and another company, perhaps forced to use an entirely different machine, offering remaining services. Ra3y Panko ------- 10-MAY-78 22:13:34-PDT,3124;000000000001 Mail from USC-ISI rcvd at 7-NOV-76 1552-PST Date: 7 NOV 1976 1536-PST Sender: STEFFERUD at USC-ISI Subject: MSGGROUP# 433 October '76 MsgGroup Transactions From: MSGGROUP at USC-ISI To: [ISI]Mailing.List: Message-ID: <[USC-ISI] 7-NOV-76 15:36:49.STEFFERUD> -- Messages from file: [USC-ISI]TRANSACTIONS.MSG;6 -- SUNDAY, NOVEMBER 7, 1976 15:32:09-PST -- 417 3 OCT MSGGROUP at USC-ISI MSGGROUP# 417 Sept '76 Transaction (5645 chrs) 418 10/06/ Pogran at MIT-Multics MSGGROUP# 418 Comment on FCC Second Computer Inquiry (2631 chrs) 419 7 OCT COMMUNICATION-DEMON a MSGGROUP# 419 Message to AV at MIT-DMS has not yet been delivered. (358 chrs) 420 7 OCT HOUGH at OFFICE-1 MSGGROUP# 420 The Second Computer Inquiry - Some Definitions (2294 chrs) 421 8 OCT DCROCKER at USC-ISI MSGGROUP# 421 Re: Comment on FCC Second Computer Inquiry (1111 chrs) 422 9 OCT PANKO at OFFICE-1 MSGGROUP# 422 Meeting on the FCC Computer Inquiry and a comment on the "Bell Bill" (1213 chrs) 423 13 Oct HENDERSON at BBN-TENE MSGGROUP# 423 Change of address from BBNA to BBNF (1237 chrs) 424 15 OCT Sles MSGGROUP# 424 Lipner now = Sles at ISIA (425 chrs) 425 18 OCT MSGGROUP at USC-ISI MSGGROUP# 425 Changes in [ISI]MAILING.LIST;93 (1625 chrs) 426 20 OCT FARBER at USC-ISI MSGGROUP# 426 Am example of the failure of systems maintainance (1255 chrs) 427 20 OCT MSGGROUP at USC-ISI MSGGROUP# 427 $Please swap Hough@Office-1 for Panko@Office-1$ (315 chrs) 428 28 OCT PANKO at OFFICE-1 MSGGROUP# 428 Murkier and murkier at the FCC (1527 chrs) 429 30 OCT MSGGROUP at USC-ISI MSGGROUP# 429 Add RCT@CCA, Replace Carlisle@ISID with Carlisle@USC-EC L (1016 chrs) 430 1 NOV MSGGROUP at USC-ISI MSGGROUP# 430 $Change Feinler@BBNB to be Feinler@OFFICE-1$ (309 chrs) 431 3 NOV PANKO at OFFICE-1 MSGGROUP# 431 Request for Information on Computer Message Services (1049 chrs) 432 5 NOV PANKO at OFFICE-1 MSGGROUP# 432 A chat with the FCC on computer mail (1994 chrs) ------- 10-MAY-78 22:13:34-PDT,315;000000000001 Mail from USC-ISI rcvd at 9-NOV-76 2209-PST Date: 9 NOV 1976 2050-PST Sender: STEFFERUD at USC-ISI Subject: MSGGROUP# 434 $Add GEdwards@OFFICE-1 & LeDuc@OFFICE-1$ From: MSGGROUP at USC-ISI To: [ISI]Mailing.List: Message-ID: <[USC-ISI] 9-NOV-76 20:50:41.STEFFERUD> Thank You, Stef ------- 10-MAY-78 22:13:34-PDT,520;000000000001 Mail from USC-ISI rcvd at 12-NOV-76 1955-PST Date: 12 NOV 1976 1938-PST Sender: STEFFERUD at USC-ISI Subject: MSGGROUP# 435 Add JZS@CCA and Dunlavey@NBS-10 From: MSGGROUP at USC-ISI To: [ISI]Mailing.List: Message-ID: <[USC-ISI]12-NOV-76 19:38:04.STEFFERUD> Please add JZS@CCA to Mailing.List: JOANNE SATTLEY, JZS@CCA COMPUTER CORP. OF AMERICA 575 TECHNOLOGY SQUARE CAMBRIDGE MA 02319 (617) 491-3670 Also, please add Dick Dunlavey@NBS-10. Thank you, Stef ------- 10-MAY-78 22:13:35-PDT,641;000000000001 Mail from OFFICE-1 rcvd at 15-NOV-76 0927-PST Date: 15 NOV 1976 0925-PST From: PANKO at OFFICE-1 Subject: MSGGROUP# 436 New Version of The Outlook for Computer Mail To: [ISI]Mailing.List: My paper, The Outlook for Computer Mail, has been updated, reduced in size, and submitted to IFIPS '77. It is about 20 pages long. If you would like a copy, send me your name, on-line address and postal address. I will send copies in two to four weeks. I would like to personally welcome Gwen Edwards (GEdwards@office-1) and Nicole Leduc (Leduc@office-1), members of the Business Planning Group at Bell Canada. ------- 10-MAY-78 22:13:35-PDT,733;000000000001 Mail from OFFICE-1 rcvd at 15-NOV-76 1437-PST Date: 15 NOV 1976 1434-PST From: PANKO at OFFICE-1 Subject: MSGGROUP# 437 A computer conference on policy issues To: [ISI]Mailing.List: NSF has funded Murray Turoff to conduct a computer teleconference on general issues affecting the future of computer teleconferencing, which, broadly defined, would include mail services. Larry Day of Bell Canada and I are chairing a session on policy issues, from December 1, 1976 through February 1977. We will be allowed to have about ten members, so I cannot invite all of you to join. But if you have a policy background (meaning regulation) and wish to take part, I'd like to talk with you about it. Ra3y ------- 10-MAY-78 22:13:35-PDT,1411;000000000001 Mail from USC-ISI rcvd at 24-NOV-76 0246-PST Date: 24 NOV 1976 0235-PST Sender: STEFFERUD at USC-ISI Subject: MSGGROUP# 438 Add Dick@ILL-NTS, introduction enclosed From: MSGGROUP at USC-ISI To: [ISI]Mailing.List: Cc: skm at NTS Message-ID: <[USC-ISI]24-NOV-76 02:35:06.STEFFERUD> I would like to find out what MsgGroup people are in- terested in. I am a social psychologist at the Center for Advanced Computation and am interested in computer networks as social systems. My group is presently under contract to the Law Enforcement Assistance Administration to design and operate a network based research support center for social scientists and state criminal justice planning agencies. We are planning for as many communication tools as the traffic will bear, and hope to form communities of clients with similar interests. I have written a paper on computer network journals and a paper on computer network support of communities of scientific researchers. Both papers are available via FTP. If they are of interest to any MsgGroup members, please tell me and I will send information on how to get them. Regards, Richard C. Roistacher Dick@ILL-NTS Assistant Professor of Sociology Center for Advanced Computation University of Illinois (217) 333-7164 ------- 10-MAY-78 22:13:35-PDT,911;000000000001 Mail from USC-ISI rcvd at 28-NOV-76 2034-PST Date: 28 NOV 1976 2026-PST Sender: STEFFERUD at USC-ISI Subject: MSGGROUP# 439 Change SMartin to be SGilbert@BBNB; remove Hornish@BBN & Gilbert@BBN From: MSGGROUP at USC-ISI To: [ISI]Mailing.List: Message-ID: <[USC-ISI]28-NOV-76 20:26:48.STEFFERUD> With congratulations we note that Shirley Martin has become SGilbert@BBNB; Also known as Mrs. John Gilbert! And, with the previous change notice, MAILER informed us that William Hornish@BBN and R.T.Gilbert@BBN no longer have network mailboxes. You may reach them at the addresses shown below. Best, Stef -------------------- R.T. Gilbert Aeronutronic Ford Western Development Laboratories Division 3939 Fabian Way Palo Alto, California 94303 Attn U-20/Mr. Gilbert William Hornish NAVOCSSACT CODE 90 Building 196 WNY Washington, D. C. 20374 ------- 10-MAY-78 22:13:35-PDT,406;000000000001 Mail from OFFICE-1 rcvd at 30-NOV-76 2129-PST Date: 30 NOV 1976 2125-PST From: PANKO at OFFICE-1 Subject: MSGGROUP# 440 Delays in sending Outlook for Computer Mail To: [ISI]Mailing.List: I haven't forgotten those of you who've asked for copies of the Outlook for Computer Mail. Printing problems have been holding up the mailing. These problems should be cleared up soon. ------- 10-MAY-78 22:13:35-PDT,541;000000000001 Mail from OFFICE-1 rcvd at 7-DEC-76 0955-PST Date: 7 DEC 1976 0930-PST From: PANKO at OFFICE-1 Subject: MSGGROUP# 441 Bert Raphael taking over as director of SRI's Augmentation Research Center, in place of Doug Engelbart To: [ISI]Mailing.List: As many of you are aware, Bert Raphael has moved over from Stanford Research Institute's Artificial Intelligence Center to become director of the Augmentation Research Center. ARC developed NLS and Journal mail under Doug Engelbart, the current director of ARC. ------- 10-MAY-78 22:13:35-PDT,5390;000000000001 Mail from OFFICE-1 rcvd at 8-DEC-76 1617-PST Date: 8 DEC 1976 1617-PST From: PANKO at OFFICE-1 Subject: MSGGROUP# 442 Carter's Use of Computer Message Service (with compliments of Ron Uhlig, who found this article) To: [ISI]Mailing.List: cc: rulifson at PARC-MAXC, taylor at PARC-MAXC, cc: sutherland at PARC-MAXC, schlesinger at PARC-MAXC, cc: metcalfe at PARC-MAXC < PANKO, CARTER.NLS;3, >, 8-DEC-76 15:17 RA3Y ;;;; .YBL=1; .YBS=2; .PN=0; .PES; COMPUTER TIED CARTER, MONDALE CAMPAIGNS The Bethesda Connection By John Holusha, Washington Star Staff Writer Jimmy Carter was sometimes described as the "computer-driven candidate" during his determined quest for the presidency. Along with computerized cost controls, the Democratic candidate had terminals humming in both his and running mate Walter Mondale's airplanes as they crisscrossed the country. It was the computer that kept track of each other's schedules and--more importantly--kept tabs on what each was saying to avoid embarrassing contradictions. If Mondale wanted to know what Carter was saying on tax reform, all he had to do was have an aide punch a few keys and the machine would come up with all his speeches on the subject. Similarly, if Carter wanted to get a message to Mondale, the fastest way was often was to punch it into the computer. As soon as the chartered campaign plane landed, it would connect up with the computer and all incoming messages would be printed out. As far as the staff on the planes were concerned, they were checking in with Atlanta headquarters. Actually, everything in the system--from the Atlanta operation to the candidates themselves anywhere in the country--was being funneled through a very powerful computer located on the third floor of the Suburban Trust building in Bethesda. The system is the brainchild of John McCann, an aggressive executive of the company that owns the computer, Scientific Time Sharing Corp. McCann is a 20-year veteran of the computer industry and a onetime amateur politician. "I had done a lot of work for Norman Cousins in the preliminary organization for the Muskie campaign four years ago," McCann says. "I knew what could be done and I had a personal interest." McCann didn't have any contacts in the Carter camp. So he started off in mid-June with a letter to Dr. Peter Bourne, who had been mentioned in press accounts as one of Carter's inner circle. He was shortly invited to Atlanta to meet with chief systems analyst Stephen Slade. "Within two weeks I had a contract." Altogether, the computerized communications system cost the campaign about $5,000. "They got a good deal," McCann said. We didn't know much about pricing so we did it on a per message basis. So they sent a relatively few long messages rather than a lot of shorter ones." McCann says the campaign workers had a strong sense of history. They wanted hard copies (computerese for something on paper) of everything for archival purposes. This system generated hard copies of all communications." The way McCann's own company uses computer communications gives some idea of what the future holds. The system is called the "Mailbox" and all an individual sees is a modern electric typewriter with some extra keys. One person in the company "deposits" a message in the mailbox by punching in a few codes and then writing out the message. It can be addressed to one other person, several or everybody on the system. Company employees check with their terminals once or twice a day to see what has been "deposited" for them. The machine prints them out in descending order of priority. The result is that face-to-face communications, old fashioned talking in the halls, is becoming rare in the company. The president might be in Bethesda, but the executive vice president is in Los Angeles and other key executives are spread across the country. "To an extent we're becoming the victims of our own copper umbilicals," McCann observes. Although he said Carter has shown clear acceptance of modern computer technology, McCann is going to go slowly in pushing it on the White House. "When that story appeared saying all the resumes were going into a computerized data bank to match them up with jobs, I heard he reacted very strongly," McCann said. "He didn't want to give the impression some robot was making decisions." McCann says he'll try to sell the White House on things similar to those that worked during the campaign: scheduling and information sorting. "Say the press office wanted some selective information, all the latest on Israel for instance. You could have hard copy almost instantly. Or if you wanted to know where the Treasury secretary was going next week and what he was going to speak on. It would all be available." So far McCann hasn't gotten any response because the Carter campaign staff has dispersed and the transition staff is just getting organized. "I'll be in touch with Carter's people," McCann says," as soon as I find out who they are." Washington Star, November 21, 1976, p. A3 ------- 10-MAY-78 22:13:36-PDT,1410;000000000001 Mail from USC-ISI rcvd at 16-DEC-76 1255-PST Date: 16 DEC 1976 1121-PST Sender: STEFFERUD at USC-ISI Subject: MSGGROUP# 443 Add DGR@SU-AI, Holg@ISIB, Turman@BBNB Subject: [DGR at SU-AI (Glenn Ricart): MSGGROUP] From: MSGGROUP at USC-ISI To: [ISI]Mailing.List: Message-ID: <[USC-ISI]16-DEC-76 11:21:36.STEFFERUD> Begin forwarded message -------------------- Mail from SU-AI rcvd at 15-DEC-76 0715-PST Date: 15 Dec 1976 0708-PST From: DGR at SU-AI (Glenn Ricart) Subject: MSGGROUP To: stefferud at USC-ISI Rob Stotz at ISI told me that you would have information on a group of people interested in machine MAIL systems.  I am the author of what I believe is the best TOPS-10 (non-ARPAnet) MAIL system, and I'm trying to collect good ideas and bad experiences before I go into the next round of design and coding on my system. My DEC-10, at the National Institutes of Health, Bethesda, Maryland, is not on the ARPAnet, and so I'm not yet aware of what's been done recently on the network. I am familiar with the mail systems sported by Stanford, Carnegie, and TENEX SNDMSG. I'd appreciate learning of any interesting documents to read on the subject. If there is an ARPAnet interaction group on MAIL systems, I'd enjoy trading ideas. . . . . Glenn Ricart ------- -------------------- End forwarded message ------- 10-MAY-78 22:13:36-PDT,430;000000000001 Mail from OFFICE-1 rcvd at 17-DEC-76 1032-PST Date: 17 DEC 1976 1008-PST From: PANKO at OFFICE-1 Subject: MSGGROUP# 444 Merry Christmas (to authorized personnel only) To: [ISI]Mailing.List: cc: day at OFFICE-1, rulifson at PARC-MAXC, cc: schlesinger at PARC-MAXC, taylor at PARC-MAXC, cc: bair at OFFICE-1 You are cordially authorized to have a merry Christmas and a happy CY77 Ra3y Panko ------- 10-MAY-78 22:13:36-PDT,1315;000000000001 Mail from OFFICE-1 rcvd at 29-DEC-76 2001-PST Date: 29 DEC 1976 1949-PST From: PANKO at OFFICE-1 Subject: MSGGROUP# 445 Request for Info on ARPANET Mail Systems To: [ISI]Mailing.List: cc: Wessler at BBN, Yonke at USC-ISI I am doing a paper that will involve a history of computer mail systems on the ARPANET. I would be grateful if some of you could answer a few questions I have. 1. When did the ARPANET become operational, and when did its planning begin? 2. When did it become possible to send messages from one host to another with a reasonable hope of success? When did the ftp, Mailer and SNDMSG features appear? 3. Can you tell me when a few of the current and past mail reading systems -- HG, RD, MSG, etc. -- were written, who wrote them, what earlier systems they drew upon, and whether they were written on project or on the side? 4. How ARPA's attitudes toward message uses on the net changed over time. When did ARPA really start to fund computer mail systems? 5. What new computer mail projects are underway or on the design board. 6. Is anybody still fooling around with teleconferencing systems on the network? 7. I would appreciate any other historical or futures trivia any of you could provide. Thank you very mcuh ------- 10-MAY-78 22:13:36-PDT,513;000000000001 Mail from OFFICE-1 rcvd at 31-DEC-76 1605-PST Date: 31 DEC 1976 1557-PST From: PANKO at OFFICE-1 Subject: MSGGROUP# 446 Responses on the ARPANET and Existing Mail Systems To: [ISI]Mailing.List: I am going on a trip for a few weeks, so I won't have a chance to thank all of you that responded to my request for information on the ARPANET and mail systems. So far the results are very interesting, and I hope to synopsize some major results for the MSGGROUP in late January. Ra3y ------- 10-MAY-78 22:13:36-PDT,674;000000000001 Mail from USC-ISI rcvd at 4-JAN-77 1724-PST Date: 4 JAN 1977 1704-PST Sender: STEFFERUD at USC-ISI Subject: MSGGROUP# 447 Add [SCHLESINGER at PARC-MAXC: msgGroup mailing list] From: MSGGROUP at USC-ISI To: [ISI]Mailing.List: Message-ID: <[USC-ISI] 4-JAN-77 17:04:28.STEFFERUD> Begin forwarded message -------------------- Mail from PARC-MAXC rcvd at 4-JAN-77 1628-PST Date: 4 JAN 1977 1628-PST From: SCHLESINGER at PARC-MAXC Subject: msgGroup mailing list To: stefferud at ISI could i be added to the msggroup mailing list.? thank you, steve ------- -------------------- End forwarded message ------- 10-MAY-78 22:13:36-PDT,920;000000000001 Mail from USC-ISI rcvd at 24-JAN-77 2121-PST Date: 24 JAN 1977 2101-PST Sender: FARBER at USC-ISI Subject: MSGGROUP# 448 Satellite Business Systems From: FARBER at USC-ISI To: [ISI]Mailing.List: Message-ID: <[USC-ISI]24-JAN-77 21:01:59.FARBER> It was noted in this weeks Aviation Week that the FCC by a 5 to 0 vote authorized the construction and operation of a specialized all digital domestic satellite system by SBS. SBS is a partnership of IBM, Aetna Casualty and Comsat General. The expected expenditure through 1985 is $400 million " The SBS system will comprise two satellites in orbit and hundreds of small ground stations with 16 foot and 23 ft antennas. It will provide integrated digital transmission including voice, data and wideband services among dispersed business locations utilizing the 12/14 gc bands " ------- 10-MAY-78 22:13:36-PDT,1254;000000000001 Mail from OFFICE-1 rcvd at 3-FEB-77 1607-PST Date: 3 FEB 1977 1401-PST From: PANKO at OFFICE-1 Subject: MSGGROUP# 449 TYMNET's ONTYM message service To: [ISI]Mailing.List: cc: taylor at PARC-MAXC, sutherland at PARC-MAXC, cc: metcalfe at PARC-MAXC, rulifson at PARC-MAXC, cc: schlesinger at PARC-MAXC TYMNET is building a computer message-switching system to be called ONTYM. It will be operational about midyear. It will be much closer to traditional teletypewriter exchange services than to computer mail systems as we know them. ONTYM will require users to prepare messages before using the system. It will then switch the message to the appropriate destimations. Average price of a message will be around a quarter, about the same as a normal private-wire switching system. The big advantage will be that the customer will be able to use existing terminals and network tie-in arrangements. Later, ONTYM will develop ARPANET-style bells and whistles. It is interesting that TYMNET believes that the main market is still in naked transmission and switching, rather than in life-cycle systems that automate not only transmsission, but aalso composition, filing and retrieval. Ra3y Panko - FYI ------- 10-MAY-78 22:13:36-PDT,699;000000000001 Mail from USC-ISI rcvd at 5-FEB-77 0121-PST Date: 5 FEB 1977 0106-PST Sender: STEFFERUD at USC-ISI Subject: MSGGROUP# 450 Add Bowerman@BBNB, WClark@BBNB, Metcalfe@PARC-MAXC & Subject: change BBNF to be BBNA From: MSGGROUP at USC-ISI To: [ISI]Mailing.List: Cc: Bowerman at BBNB, WClark at BBNB, Metcalfe at PARC-MAXC Message-ID: <[USC-ISI] 5-FEB-77 01:06:30.STEFFERUD> Welcome to Tom Bowerman@BBNB and Tom McLaughlin (WClark@BBNB), and Bob Metcalfe@PARC-MAXC. Also, we note that BBN has changed their internal accounts and mailboxes around again, so you should reset BBNF to be BBNA for Burchfiel, Henderson, Mooers, Myer, and Vittal. Best regards, Stef ------- 10-MAY-78 22:13:37-PDT,789;000000000001 Mail from USC-ISI rcvd at 5-FEB-77 0904-PST Date: 5 FEB 1977 0855-PST Sender: FARBER at USC-ISI Subject: MSGGROUP# 451 Paper on computing and communications From: FARBER at USC-ISI To: [ISI]Mailing.List: Message-ID: <[USC-ISI] 5-FEB-77 08:55:58.FARBER> There is the text of an invited paper by Paul Baran and me for AAAS SCIENCE entitled " The Convergence of Computing and Telecommunications systems" available in [ISIA]Science.paper. This text is essentially that to be published except for a few grammatical changes that have not been included in the on-line version. If you have any problems with getting the paper via the net , sndmsg and I will either sndmsg or US Mail you a copy. Dave ------- 10-MAY-78 22:13:37-PDT,324;000000000001 Mail from USC-ISI rcvd at 5-FEB-77 2324-PST Date: 5 FEB 1977 2312-PST Sender: STEFFERUD at USC-ISI Subject: MSGGROUP# 452 $Please Delete Sutkowski@OFFICE-1 from Mailing.List$ From: MSGGROUP at USC-ISI To: [ISI]Mailing.List: Message-ID: <[USC-ISI] 5-FEB-77 23:12:17.STEFFERUD> Thanks, Stef ------- 10-MAY-78 22:13:37-PDT,1097;000000000001 Mail from BBN-TENEXB rcvd at 8-FEB-77 0617-PST Date: 8 Feb 1977 0843-EST Sender: BOWERMAN at BBN-TENEXB Subject: MSGGROUP# 453 introduction [bowerman@bbnb] From: dmis,anad To: [ISI]Mailing.List: Message-ID: <[BBN-TENEXB] 8-Feb-77 08:43:50.BOWERMAN> Keywords: introduction greetings: i will keep this introduction short. i am a customer and a user. i am the director of management information systems at Anniston Army Depot. we have three directories on bbnb at anniston-all in mgt info sys. however, we will have another one up for the commander of the depot on 1 march 77. i am interested in the various message systems in use and prefer HERMES over all others i have used. i am also interested in the other features available. i am not the expert that most of you are, but i am a good listener and willing to give feedback from a user standpoint when specifically requested. i have read every msggroup message created since day one. you are performing a valuable service. regards tom ------- 10-MAY-78 22:13:37-PDT,1576;000000000001 Mail from USC-ISI rcvd at 15-FEB-77 0031-PST Date: 15 FEB 1977 0027-PST Sender: STEFFERUD at USC-ISI Subject: MSGGROUP# 454 [ISI] directory listing From: MSGGROUP at USC-ISI To: [ISI]Mailing.List: Message-ID: <[USC-ISI]15-FEB-77 00:27:08.STEFFERUD> To facilitate FTP of the Proceedings files, they have been broken up as shown in the following directory listing. #nnn-#mmm indicates the range of MSGGROUP# nnn messages in each file. To obtain a copy of any of these files via SNDMSG, send a request to MsgGroup@ISI. Best Regards, Stef [USC-ISI] 15-FEB-77 00:12:45 PGS MESSAGE.TXT;1;P774444 5 ADMINISTRATIVE.MSG;9;P777752 53 .SURVEY;6;P777752 8 PROCEEDINGS.MSG#001-#050;1;P777752 38 .MSG#051-#100;1;P777752 26 .MSG#101-#150;1;P777752 60 .MSG#151-#200;1;P777752 63 .MSG#201-#250;1;P777752 47 .MSG#251-#300;1;P777752 47 .MSG#301-#350;1;P777752 27 .MSG#351-#400;1;P777752 24 .MSG#401-#450;1;P777752 24 .SURVEY#001-#050;1;P777752 2 .SURVEY#051-#100;1;P777752 2 .SURVEY#101-#150;1;P777752 3 .SURVEY#151-#200;1;P777752 2 .SURVEY#201-#250;1;P777752 3 .SURVEY#251-#300;1;P777752 3 .SURVEY#301-#350;1;P777752 2 .SURVEY#351-#400;1;P777752 3 .SURVEY#401-#450;1;P777752 3 TRANSACTIONS.MSG;8;P777752 410 .SURVEY;8;P777752 27 MAILING.LIST;107;P774646 1 RECIPIENTS.MAILING-LIST;5;P777752 1 ------- 10-MAY-78 22:13:37-PDT,1085;000000000001 Mail from OFFICE-1 rcvd at 16-FEB-77 1045-PST Date: 16 FEB 1977 0947-PST From: PANKO at OFFICE-1 Subject: MSGGROUP# 455 CB Computer Mail Draft To: [ISI]Mailing.List: cc: rulifson at PARC-MAXC, taylor at PARC-MAXC, cc: sutherland at PARC-MAXC, metcalfe at PARC-MAXC, cc: schlesinger at PARC-MAXC, carlisle at USC-ECL, cc: martin at USC-ECL I am writing a paper suggesting that "CB Computer Mail" -- the combination of hobby computing and message technology -- may be the first really big manifestation of computer mail. The paper is half-whimsical, half-serious. It is motivated by the explosive growth of CB radio and the healthy growth of hobby programming. Before you reject the notion out of hand, it should be noted that ARPANET programmers tend to be hobby computerists -- albeit on a grand scale. If you are interested in reviewing the draft, it is at office-1, in the panko directory. A standard text version is entitled FTP-CB. An NLS version is named PUBLIC-CB. I would appreciate your comments. Ra3y Panko SRI ------- 10-MAY-78 22:13:37-PDT,13009;000000000001 Mail from OFFICE-1 rcvd at 16-FEB-77 1617-PST Date: 16 FEB 1977 1609-PST From: PANKO at OFFICE-1 Subject: MSGGROUP# 456 CB Computer Mail Draft To: msggroup at USC-ISI, farber at RAND-UNIX, To: schlesinger at PARC-MAXC < PANKO, MULE.NLS;1, >, 16-FEB-77 09:25 RA3Y ;;;; .PN=0; .PES; .YBS=3; .YBL=2; .F="February 1976 .Split; Page .GPN;"; .Center=13;.YBL=0;.GCR=5;CB COMPUTER MAIL? by Raymond R. Panko Stanford Research Institute Menlo Park, California 94025.YBL=1;.GCR=3; 1 During his recent election campaign, President Jimmy Carter coordinated his travelling staff with his headquarters people by means of a new communication medium called "computer mail." Whenever Carter's campaign jet landed, a staff member would call into the nationwide STSC computer network and call up a program called "mailbox." During the next few minutes, he or she would receive messages sent hours or seconds earlier, send out messages prepared on the plane, and perhaps type back brief replies. 2 Unlike standard teletypewriter services (such as Telex and TWX), Mailbox did not forget messages once they were delivered. It filed important messages away, so that Carter's staff could refer to them later. Furthermore, Mailbox made it easy for people to compose, send, and read their mail -- thus eliminating the traditional Telex/TWX priesthood tha normally stands between users and the system. In essence, Mailbox combined modern word processing computer tools with the instant delivery of teletypewriter networks. The result was a system simple enough for anyone to use, yet powerful enough to automate virtually every part of the message handling process, from initial composition through final filing and retrieval. 3 Computer mail is a product of this decade. Although it can trace its roots to the crude mailbox programs on early time shared systems and has some kinship to the mailbox tools on most commercial networks, such systems have only pale resemblances to computer mail. For the most part, computer mail was developed after 1972 on the ARPANET. The ARPANET a the computer network run by the U.S. Defense Communications Agency. It links most major computer science research centers in the United States. Originally, the ARPANET was created by the Defense Advanced Research and Development Agency (ARPA), which funded much of the advanced computer research in the United States during the 1960s. The ARPANET was originally designed to let ARPA contractors share resources on distant computers. But some ARPANET users soon developed software for message handling, and soon mail systems grew like wild fire, borrowing features from one another and evolving at a rapid rate. 4 Before 1975, most computer mail development on the ARPANET was informal and was often bootlegged onto other projects or written in users' spare time. But the value of computer mail had become obvious to ARPA by the bigenning of 1975. ARPA, like a number of other nonprogramming organizations, had begun to use computer mail for its bread and butter communications, and had become aware that a relatively mature communication medium was becomming available. After 1975, ARPA took a more active role in computer mail, formaing two groups that were to help devise network standards for mail systems, and interesting the Navy in an operational test of computer mail. For the Navy Military Message-Handling Experiment, ARPA funded three organizations to develop advanced computer mail systems. This year, at least one of these systems will be tested in a Naval communication center in Hawaii. 5 While computer mail was flourishing on the ARPANET, a parallel development was taking place primarily through National Science Foundation (NSF) and ARPA funding. This was computer teleconferencing. In teleconferencing, the computer enforces communication dynamics that resemble communication around a conference table. Computer teleconferencing began around 1971, when Murray Turoff built a conferencing system for the Office of Emergency Preparedness. Since then, the FORUM conferencing system (whose commercial variant is called PLANET) designed at the Institute for the Future and the EIES system developed by Turoff at the New Jersey Institute of Technology have advanced the state of the art considerably. Most conferencing and computer mail designers believe that the two design trends will eventually merge into a more complex computer medium. For simplicity, we will refer to future computer media as computer mail. 6 The commercial sector has generally lagged behind the ARPANET and conferencing communities by several years, but it too shows signs of growing interest in computer mail. The service that Carter used was a commercial service halfway in sophistication and power between standard commercial services and ARPANET services. In addition, one of the computer mail systems developed for the Hawaii test, HERMES, has recently been made available on a commercial computer transmission network, Telenet; and PLANET was made commercially available, on a restricted basis, in 1974. 7 Service sophistication is one thing but cost is another. The three-minute telephone call is a very cheap way to communicate, especially when the communicator's time is taken into account. Unless computer mail can offer low prices, it will remain a presidential service, or at best a service for hothouse environments like NSF experiments, the military, or the ARPANET. At present, computer mail is reasonably expensive, although not prohibitively so. A typical message (which averages only about 100 words on most systems) costs about $1.50 to compose, transmit, print, file, and retireve. A few new systems are more expensive than this, costing over $4.00 per message. Overall, it is now possible to build a computer mail system that is no more expensive than Telex or TWX yet offers much more. Since the Telex/TWX market is over 100 million messages annually in the U.S., and since private corporate teletypewriter networks carry far more traffic than this, there is a potentially large market available, even at current prices. 8 But it is future prices that catches the eye. Computer costs are falling rapidly, and if intelligent terminals can be used to reduce communication costs, it should be possible to compose, send, and receive a typical message for around 25 cents. This is the conclusion of two recent SRI studies by the author (a,b). Since the cost of postage has been projected at 23 cents for 1980 (c) and as much as 35 cents (d) for 1985, it seems that computer mail will be cheaper than postage, despite automating many more message functions. While any cost projection made at this time must be highly tentative, it does appear that computer mail may become an important new communication medium. One of the author's studies (b) found that the total volume of written communication in the United States may exceed 200 billion messages annually, and this may be a low estimate. If computer mail can compete not only for postal message delivery, but also for interoffice mail, intraoffice mail, and telephone communication, it may be one of this country's largest industries during the decades ahead. 9 Most speculation about computer mail has tacitly or explicitly assumed that the next surge of computer media will come in the corporate sector. Corporations, it has been argued, can afford to pay the initial costs of computer mail development and use. In addition, corporations have the communication density to support initial applications. To most analysts, home uses seem far away. 10 But three developments should give us pause. The first development is negative: commercial users' inability, so far, to perceive the advantages of computer mail. In commercial computing, the emphasis is on cheapness and conservativeness; many new communiction services are finding hard sledding in this environment. The second development is positive: the spectacular growth of CB radio in the United States. There are around 15 million CB's in the United States today, as compared with only 70 million households. People have found that they love to communicate in the party-line CB environment, which offers socializing yet, if desired, annonymity. The third development is the more modest but equally exciting flowering of hobby computing. 11 It sseems distinctly possible that the 15,000-odd hobby computerists in the United States today [3] have all the characteristics needed to create a "CB Computer Mail" phenomenon: the expertise needed to put together message software on local computers (for cheap interfacing with the central computer), a strong sense of community and interest, and a willingness to innovate. It seems possible that if some corporation were to install a central computer media computer in any hobby computer hotbed, such as Los Angeles, San Francisco, or Cambridge, there would immediately be a healthy volume of traffic in both mail and teleconferencing modes. 12 Let me offer several specific reasons for suggesting that hobby computerists may be the first big users of computer media. Historically, computer media were first extensively developed on the ARPANET. Anyone familiar with the Advanced Research Projects Agency (after whom the ARPANET is named) realizes that ARPA was the dominant funder of leading-edge computing during the 1960's. Essentially, ARPA was funding the community of hobby computerists par excelence. Funding was fat and creativity was given free reign during business hours. Moreover, ARPA contractors found their staffs working long overtime, developing space war games, stock market information services, and--as noted above, computer mail systems. In other words, hobby computing at a grand scale was the original source of many advanced mail systems today. Computer mail had a strong hobbiest flavor in its use as well as in its origins. Colleagues in artificial intelligence, data base design, and other exotic fields used comuter mail to build and maintain their community. 13 Furthermore, in applications where computer teleconferencing has been successful, discussion has often been free-wheeling and chatty. The longest conferences tend to be breezy and rambling, yet very successful in exchanging ideas and viewpoints. Nowhere is this party-line flavor more noticeable than in conferencing systems with "grafitti" conferences open to all members. According to Gordon Thompson of Bell Canada's Bell Northern Research Institute, sttatistics from Bell Canada's CMI-3 computer conferencing system were dominated by the graffiti conference [4]. 14 This is how a hobby computer media network might develop. Someone would write compatible mail packages for several hobby computers and for a central computer, then drop the computer in a hobbiest hotbed. Hobbiests would then access the computer through simple terminals or through their hobby computers. There would be open discussions like bulletin boards, conferences for special interest groups, and probably an on-line yellow pages (a la a similar service on the ARPANET) to announce new products. There would also be a mail service for one-to-few interactions. 15 The hang-up in all of this, of course, is billing the hobbiest users. One alternative would be for a few major vendors to set up, in effect, an on-line byte magazine. Another would be for a club to set up a very simple and cheap message switching computer and charge members on a per-use basis, with the threat of service cancellation or even membership cancellation, in case of nonpayment. A third scenario would call for a major computer company to fund a system to probe the market for hobby computer use, just as Weyerhouser funded counter-culture communities in the 1960's; those experiments persuaded Weyerhouser to enter the indoor plant market aggressively, a decision which paid for the original experiments many times over. 16 ------- 10-MAY-78 22:13:38-PDT,1299;000000000001 Mail from OFFICE-1 rcvd at 16-FEB-77 1651-PST Date: 16 FEB 1977 1629-PST From: PANKO at OFFICE-1 Subject: MSGGROUP# 457 Oops. The directions I gave for getting the CB paper won't work! To: [ISI]Mailing.List: or< PANKO, MESS.NLS;1, >, 16-FEB-77 16:14 RA3Y ;;;; As about 200 of you have already learned, you cannot get either the NLS or Tenex versions of the CB Computer mail paper using the directions I gave in the previous message. For one thing, I told you that the Tenex version is TFP-CB. It is actually FTP-CB.txt. I didn't know that you had to know the extension to do an FTP across the net. As to the NLS version, which is Public-CB.nls, I really don't know what the problem has been. Some people have accessed it just fine, while others report problems. I honestly don't understand NLS privacy protection. At any rate, I will be messaging a version to MSGGROUP at ISIA. Stef will give it a number, and you can access it there. I am terribly sorry about this mix-up. I wish message systems could handle long messages more easily. As usual, Ken Pogran of MIT has already made one memorable comment about the idea of CB Computer Mail. He suggests that we will have to look out for the Data Smokies. Is that the FCC, Ken? ------- 10-MAY-78 22:13:38-PDT,431;000000000001 Mail from USC-ISI rcvd at 16-FEB-77 2204-PST Date: 16 FEB 1977 2151-PST Sender: STEFFERUD at USC-ISI Subject: MSGGROUP# 458 CB mail From: Farber at Rand-Unix To: panko at OFFICE-1 Cc: [ISI]Mailing.List: Message-ID: <[USC-ISI]16-FEB-77 21:51:15.STEFFERUD> just a query. Is it legal to transmit ASCII over CB. It is not yet legal over ham if I remember right. Old baudot is though. Hmmm Dave ------- 10-MAY-78 22:13:38-PDT,890;000000000001 Mail from USC-ISI rcvd at 16-FEB-77 1946-PST Date: 16 FEB 1977 1942-PST Sender: STEFFERUD at USC-ISI Subject: MSGGROUP# 459 Re: MSGGROUP# 456 CB Computer Mail Draft From: MSGGROUP at USC-ISI To: Panko at OFFICE-1 Cc: [ISI]Mailing.List: Message-ID: <[USC-ISI]16-FEB-77 19:42:32.STEFFERUD> In-Reply-To: Your message of FEBRUARY 16, 1977 Hi Ra3y, Your CB Computer Mail Draft is now comfortably parked in a file by itself in [ISI] as shown below. Anyone can get a copy by using FTP as described in the MsgGroup procedures, or by requesting a copy by SNDMSG to MSGGROUP@ISI. Enjoy, Stef -- Messages from file: [USC-ISI]MSGGROUP#.456;1 -- WEDNESDAY, FEBRUARY 16, 1977 19:12:40-PST -- 1 16 FEB PANKO at OFFICE-1 MSGGROUP# 456 CB Computer Mail Draft (13009 chrs) ------- 10-MAY-78 22:13:39-PDT,2031;000000000001 Mail from USC-ISI rcvd at 16-FEB-77 2231-PST Date: 16 FEB 1977 2217-PST Sender: STEFFERUD at USC-ISI Subject: MSGGROUP# 460 Request for Voluntary Introductions From: STEFFERUD at USC-ISI To: [ISI]Mailing.List: Message-ID: <[USC-ISI]16-FEB-77 22:17:42.STEFFERUD> Start-Date: 15 SEP 2173 End-Date: 15 SEP 2173 Suspense-Date: 15 SEP 2173 A long time ago we asked MsgGroup members to send us a brief introduction of themselves for the benefit of other members, and we actually received a few. (MsgGroup# 225, 232, 236, 250, 309, 416, 453) Recently I have received requests from a number of people (not all new members) for such information. Thus, I have set up a file [ISI]INTRODUCTIONS.MSG in which we will keep the latest version of any introductions we receive from members. It is suggested that we not clog each other's mailboxes with a flood of these introductions. Please just send single copies to MsgGroup@ISI and I will file them in INTRODUCTIONS.MSG, and send out a survey of the introductions as they accumulate. And, you can FTP the file to look at it or request copies as you may wish. What we would like is for you to send a very brief description of yourself to the MsgGroup so we will know what your first names are, where you work, what position you hold (title?), and a bit about what role you play in this drama. For example, are you: 1. a customer (one who makes budget commitments to buy message system services), 2. a user (one who uses message system services purchased by someone else), 3. an implementer (one who builds message systems), 4. a sponsor (one who funds both users and implementers in the development effort), 5. an analyst (one who analyzes and evaluates systems for a fee), 6. a critic (one with no vested interest who contributes constructive comments), or 7. something else? See you in the discussions, Stef ------- 10-MAY-78 22:13:39-PDT,879;000000000001 Mail from RAND-UNIX rcvd at 17-FEB-77 0925-PST Date: 17 Feb 1977 at 0922-PST Subject: MSGGROUP# 461 INTRODUCTION from Stock Gaines@RAND-UNIX Subject: Re: MSGGROUP# 460 Request for Voluntary Introductions From: Gaines at Rand-Unix To: MsgGroup at Usc-Isi Message-ID: <[Rand-Unix]17-Feb-77 09:22:18.Gaines> In-Reply-To: Your Message of 16 FEB 1977 2217-PST In-Reply-To: <[USC-ISI]16-FEB-77 22:17:42.STEFFERUD> I am a member of the Information Sciences Dept. at Rand. I am a recent convert to the use of mail systems. I am interested in the evolution of ideas about electronic mail, and more generally in communications via computer as a major aspect of new interactive uses of computers. I am also interested in the system questions which arise here, such as process structuring questions, file and directory system questions, etc. Stock Gaines ------- 10-MAY-78 22:13:39-PDT,34075;000000000001 Mail from USC-ISI rcvd at 17-FEB-77 1047-PST Date: 17 FEB 1977 1039-PST Sender: FARBER at USC-ISI Subject: MSGGROUP# 462 An Invited paper for a special issue of AAAS SCIENCE Subject: journal on Electronic. 'The Convergence of Computing Subject: and Telecommunications Systems' by Dave Farber and Paul Subject: Baran From: FARBER at USC-ISI To: MSGGROUP Message-ID: <[USC-ISI]17-FEB-77 10:39:08.FARBER> THE CONVERGENCE OF COMPUTING AND TELECOMMUNICATIONS SYSTEMS by David Farber University of California at Irvine Information and Computer Science Irvine, California 92717 (714) 833-6891 and Paul Baran Cabledata Associates 701 Welsh Road Palo Alto, California (415) 328-2411 I. PREFACE A major change in computer communications is taking place. As the cost of communications and computation declines, the number of new services that can be offered will increase and a large number of new industries can be expected to emerge. A battle between the potential suppliers is now underway. This paper considers: 1. the nature of the new services, 2. their importance to the economy, 3. the arguments for regulatory control as proposed by the major potential competing suppliers. The basis for seeking regulation hinges mostly on the use of a communication component provided by a historically highly regulated natural monopoly. Rapid changes in the technological basis make it difficult, if not impossible, to resolve the differences between the three major competing suppliers using the existing communications regulatory framework. The concept that computation and communications are distinctly separable areas no longer resolves boundary disputes rapidly, when regulatory issues are involved. This paper examines the underlying technical changes, reflected in the conceptual changes, which now make the regulators' tasks difficult. It suggests that part of the present turmoil and uncertainty is the growing use of digital processes within the communications industry itself - an inevitable evolution of the new technology. A tutorial is appended to describe the evolution of data transmission technology. We do not intend to propose solutions to the very difficult problems that this paper raises. We do not know of any completely workable solutions. Rather, we intend to sensitize the reader to the issues and problems that our country will face now and in the future. II. The Nature of the New Services Today - An Example This paper was written by two authors 500 miles apart. Each writes on the same sheet of "paper". Each adds, modifies and clarifies the words of the other. Misspellings are corrected and the order of phrases changed. Each author takes his turn tidying up the modified manuscript by pressing a few buttons and giving a few commands. This takes care of the nitty gritty in paper writing, gobbling up the blank spaces and even justifying the right margin. Discussions on content and outlooks take place using the same computer communications based message systems. This paper you are currently reading is not tomorrow. It is today; now. Tomorrow, computer communications systems will be the rule for remote collaboration, like the above example . As computer communication systems become more powerful, more humane, more forgiving and above all, cheaper, they will become ubiquitous. The obvious next question is, "Why bother to retype this manuscript manually, set type and print this journal on paper?" Once these words have been entered into the computer system they are in fact "published" if a copy has been stored in a publically accessible memory portion of the computer communication system. Think of the waste of paper in conventional periodical publications when only a small portion will ever be read. The cost component of the computer communications cost/demand curve declines each year. Yet, there are literally orders of magnitude more improvement still possible from what is known already in the laboratories. Accelerated expansion of demand is a certainty. With increased use of computer communications, changes in our institutions will follow. The example given for publishing is only suggestive of the range of change. The immediate reaction of those in publishing to such change is that of threat. Institutions rarely take kindly to the prospect of obsolescence of their present justification. III. IMPORTANCE TO THE ECONOMY Our simple illustration of the potential change in publishing could, and most likely will in time, spread across the entire economy. Even today we are an information processing society. More Gross National Product (GNP) is developed in the US today in information sector activities than in the production of tangible goods. We have chosen a dramatic application of computer communications solely for illustration. More important, are the mundane applications: automated hotel reservations, credit checking, real-time financial transactions, access to insurance and medical records, general information retrieval, real-time inventory control in business, etc. A wide range of services such as these and beyond is coming. Automation of the office is almost a certainty. The only questions are "when," "how," and "by whom?" Electronic transmission of much of the current First Class Postal mail similarly will follow. Change will be felt widely throughout the economy and, in a fundamental manner, by many institutions. Such change is generally perceived as crisis and rarely accepted with grace. Change is something best absorbed by others. The interim period will be a period of inevitable discomfort to those organizations and professions which are affected. IV. THE ARGUMENT FOR REGULATION History of Potential Suppliers in this Information Economy While computer communications promises to provide huge new markets for potential suppliers, there is uncertainty as to who they will be. It is not simply a choice of AT&T, IBM or a diversity of smaller computer companies. Rather, the question is which supplier will capture the largest, most profitable slice of the pie, and with the minimum degree of the unpleasantness of competition. Each organization arrives at its claim of legitimacy to supply the new markets by a different route. The telephone company supplies communications and is now a regulated monopoly. ( Providing basic communication channels is most economically supplied on a common shared resource basis -- it is almost unarguably a "natural monopoly.") IBM's claim in this new industry is well founded. It will provide a significant share of the new market's hardware. It is cash rich and thus unique in the computer industry in having the financial resources to make the needed capital investments. And, it is the winner in the proof-by-adversary test in a free economy. It has been tried and has won in the free marketplace. The third contender, the remainder of the computer industry, feels that IBM has won unfairly and seeks to handicap the giant. (There are few computer companies today that have not brought suit against IBM for alleged violations of anti-competitive statutes.) These smaller companies feel that IBM may exercise monopolistic power in a sector in part deemed not to be a natural monopoly and thus not regulated. This may all seem strange in a free economy. But, the regulatory and institutional constraints are so treacherous that the outcome of the debate may hinge more on political adroitness than market acceptance. The Disintegration of the Boundary between Suppliers The basic premise that communications functions and computational functions are separable has been rendered meaningless by a basic and pervasive change of the technology. The separate atoms of communications and computers form a new molecule with properties different than their atomic components. Consider the recent case in which AT&T wished to offer its subscribers the sophisticated Model 40 communications terminals made by its wholly-owned subsidiary, Teletype Corp. A series of delays ensued as competitors argued that AT&T was entering the processing business. AT&T said "nonsense, it is just in the communications business,". The Common Carrier Bureau of the Federal Communications Commission decided against AT&T's position. After a short delay, the Commissioners overturned their own Bureau. Indecisiveness, on-again, off-again decisions at this boundary line are common. There is no longer an unambiguous dividing line upon which to reach decisions. The lawyers involved in resolving these arguments are assured of an employment stability lacking in most other sectors of the economy, as the endless suits at the interface between suppliers and the government seem to be careers rather than occasions. The Cause of the Disintegration The communications regulatory paradigm assumes that communications is communications, and computing is computing. Unlike oil and water they now mix well - too well to treat separately, once combined. They simply are no longer separable into orthogonal components capable of different legal treatment. Separating communications and computational processes is becoming an almost nonsense activity, like unscrambling an omelet. It is a task better left to philosophers than to regulatory commission lawyers. This has not always been the case. The basic changes have crept upon us gradually. The forces melding these two previously separable domains are several. At a gross level, there is the economics of distributed processing -- it may be cheaper in the future to provide independent computing power for those who need it - at their terminal- at their office or home. But just having one's own computer is in general not adequate. There is the occasional or possibly frequent need for shared information where, irrespective of cost, access to remote shared data in real time is mandatory. But above all is the basic nature of the communications process in which digital bits provide the natural common form for all signal transmission and processing. The digital waveform becomes the universal package with which to send voice, data, facsimile or anything else. Why this is so, and how it got that way, provides insight to the evolving technology and the increasing difficulty that will be felt by those who try to reach decisions on the premise that communications are separable in computer communications networks. What are the alternatives available for solving our dilemma. Why Aren't Microcomputers a Perfect Substitute for Large Interconnected Computers? There is a school of thought that and argues that computer communications is a chimera. There is no need to use the resources of the telephone company or IBM. "Computers are going to be so small and cheap in the future that we don't need to use communications lines to talk with big computers." Is the mini- micro-pico low cost computer the answer? Powerful computers have become cheap and will get cheaper. The explosion of the computer hobby market during the last year and a half has been remarkable. There are now 150 stores selling low cost computers in hobby kit form. More computer development work is being done by hobbyists today than by computer science research funded by public agencies . The hobbyists are not all amateurs. They are those whose vocation is also their avocation. They are smart, dedicated persons no longer blocked from entry into computer research by the cost of the equipment. And, there are many more now at work (or play?) than the government could possibly support. The reasons that access to communications capability will always be a necessary adjunct to the small computer are basically unchanged: 1) Whenever it is necessary to make infrequent use of a highly specialized resource, it is cheaper to share (why build a machine to make nails when you can buy the few you need at the corner hardware store). 2) The output of one data processing process is usually the input for another process. 3) By agglomerating computing power by interconnection of a number of previously independent systems, applications that are currently infeasible in the data processing area can be handled. These include a) the need for high availability environments, i.e. extreme reliability, b) systems that can grow modularly as their environments grow (to avoid the N+1 problem -- that is the problem of what to do with a system that supports N users and suddenly gets the N+1 st. user) , and c) allowing the physical organization of computing power to match the corporate organization , thus relieving organizational strains and stresses. Distributed Processing Technology and its Fulfillment All of the above points can be satisfied via the route of distributed processing. Distributed processing is loosely defined as the use of a multiplicity of computational devices for a single processing task. Distributed processing implies that the units that compose the distributed computer can freely intercommunicate. If the distribution is other than local, then remote computer communications facilities are required. Data Processing Begets Data Processing Consider the following example suggesting the nature of distributed processing. A grocery cash register -- really a counter computer -- computes a customer's bill quickly. But, it also generates the information needed by the store to order fresh turkeys today for delivery tomorrow. This same information can be used by a hypothetical farmer to know instantly whether there will be a shortage or a surplus of a product, and whether to hold out for a better price or sell his turkeys. This information has economic value to his procurement decision. The key feedback signal of a free economy in the past has been the flow of money. More rapid signals can be obtained as an inexpensive by-product of the flow of information from other processes. Economic examples are not the only use for computer networking and distributed processing. Equally, or more significant, are those that deal with human communications. Just as the authors of this paper were able to interchange ideas and manuscripts over the computer communications media interactively, so will users of the low cost computer systems of the future want to intercommunicate for ideas, pleasure and business. One has only to look at the meteoric growth of Citizen's Band radio to realize new opportunities for communications between people in our society. Near-zero cost micro machines making efficient use of a shared communications resource can open up entirely new forms of human communications. The Digital Channel - the Evolving Choice for Communications Transmission Before considering this and other issues in detail, it is helpful to review some of the characteristics of the digital channel and the reasons why it is becoming desirable even for the communication of basically analog signals. An analog signal is a voltage sent over the usual telephone line representative of (or an analog of) the pressure waveform of the human voice. Why should such a conceptually simple waveform be translated into a series of bits? Simply because once converted to digital format, it is possible for the waveform to tolerate a tremendous amount of distortion of its waveshape and still be capable of simple reconstruction into a neat distortion- free series of pulses. This allows greater economy of packing information in a communications circuit and allows the use of lower quality (and cost) channels. Once in digital format, pulses that have been lost or converted into the wrong symbols are recoverable. It is relatively easy to perform many rapid calculations upon the data stream to detect and correct errors. (The price is a small cost for a few extra bits to serve as error detectors and correctors.) The ultimate in such distortion removal is cryptography, where a bit stream is purposely modified by a known-in- advance bit stream (called the key). One "subtracts" one binary waveform from the other and reconstructs the initial waveshape without distortion of the transmitted intelligence. The eavesdropper sees only a binary stream with a seemingly random chance of either being a "1" or a "0". "1's" and "0's" are the names of the two states possible in the digital channel. Affordable secure transmission mandates digital transmission. The digital stream simplifies the transmission design task. Again, if you keep the signal from becoming too distorted, below 50% of its initial value, it can always be reconstructed without errors. Digital modulation, unlike analog signals which require critical adjustments merely to minimize distortion, is thus the ideal waveshape for transmission. Switching Communication networks require both transmission and switching. Switching is the process of connecting any network user to any other. Switching, by its nature, is a digital process. In the 1930's, an operator plugged one circuit into the next of a tandem chain of connections through telephone switching offices. This is a digital process as each switch is either open or closed. A neat binary process.The decision process of laying out switches and relays for telephony was among the earliest practical uses for Boolean or binary logic. The foundation of the logical design process in computer design owes much to the telephone industry. And, the integrated circuit devices which make real time computer applications feasible are all offsprings of the transistor developed at the Bell Telephone laboratories. The telephone dial itself is a digital device. A series of pulses from the dial acts to set switches. As the computer technology matured to the point where its products were realizable as well as elegant and complex, the quaint electromechanical computers that formed the telephone switching network of the past were replaced by computer circuit logic to open and close the telephone switches.This evolution continues as digital transmission is now being combined with digital switching. The two separate processes in time are melding into a single integrated process. Bits going into one end of the network are thus beginning to be treated almost the same as a bit stream in a computer, albeit with less processing. WHAT IS THE REGULATORY ISSUE Who Should Be The Future Legitimate Supplier of the Digital System? As the telephone technology slowly evolves towards an almost all-digital system in the far future, it will or can develop the technology needed to provide many of the new computer communications based services. AT&T has, for example, already developed devices for credit card transactions and sophisticated computer input/output terminals through its wholly- owned Teletype Corporation. Equally comfortably in the domain of processing and perhaps even transmission is the highly competitive computer industry. During the last decade , the FCC encouraged competition in transmission for other than voice telephone services. New transmission companies and independent "value-added carriers" were authorized. The value-added carriers essentially lease telephone channels from the communications carriers, add processing, and use the leased lines so effectively that they are able to sell the joint product for a profit over and above the cost of the basic regulated channel. Today, much of the in-fighting is via legal and regulatory actions seeking to block new service offerings and capabilities. Partially it is an attempt to prevent the opponents from establishing beachheads by precedence. And, partially, it is a way of adding uncertainty to the industry to discourage capital investment. Any new computer communications service offered today by a potential supplier is certain to be opposed by the other major contenders. IBM's plan for a communications satellite network for major business users, which bypasses AT&T Long Lines, faces opposition from both the telephone company and computer companies fearful of IBM's assumed long range intent. AT&T's offering for sophisticated terminals is met with countervailing opposition from computer terminal suppliers. The outcome of the competition between the suppliers will be decided, in part, by the staying power, legal resources and political muscle each allocates to the fray. Little attention is paid to the "public interest". In part, the term defies definition. Is the public interest the interest of the cross-subsidized residential telephone user? Is it the interest of a business which faces a reduced communications bill? Is public interest to be viewed primarily in the short term irrespective of long term damage to existing institutions in achieving immediate savings? There is no simple calculus to reach decisions. At present there is a struggle between the proponents of the advantages of competition versus those who hold that better use of limited capital results by virtue of monopoly. It is unclear which view will prevail. Until it is, there will clearly be resources wasted and future economically beneficial services to the nation delayed simply by the inability of industries to cooperate. SUMMARY Any paper that describes a problem issue is expected to have a grand finale, preferably with a brilliant solution. Were it only that simple. Lacking any elegant answer, we shall settle for a few general observations: 1. The public availability of socially usefully computer communications services is and has been held back by legal battles that are now underway by the potential suppliers. 2. No simple resolution of these issues in the near term seem likely using the past conceptual separation of computers and communications doctrines. 3. The current policy is to determine whether the nation shall or shall not have certain computer communications services by the adversary process. In this process, often only the voices of the loudest adversary suppliers are heard. 4. While there can be no certainty that better alternatives cannot be devised, we believe that such a possibility assumes a higher probability if the key actors came from the technical community sectors more representative of the future consumers. 5. If we are to have the new services that are possible, we need an approach that makes better use of the technologists dreams and goals rather than having future prospects excessively bound by lawyers paid to preserve their paid clients interests irrespective of any secondary consequences. 6. We cannot be sanguine about this possibility as technological statesmanship is too easily corrupted by the same forces that have placed us in this predicament. Further, even if not corrupted, beneficial cooperation can too readily be regarded as simply collusion. 7. In summary, while we do not have any clear answer, we do know that present approaches are not taking us where we want to go very rapidly and that alternative approaches should at least be considered. APPENDIX TECHNICAL ADVANCES IN THE DIGITAL COMPUTER COMMUNICATIONS TECHNOLOGY Below we consider the technical evolution of digital communications. Modulation The first digital transmission circuit was the electrical telegraph, predating the voice telephone by 35 years. While local telephone lines can and are used for simple on/off direct current signals, the need to transmit signals beyond the limited distances where a direct current path exists caused the use of tone or alternating current transmission. The device that converts the binary signal into an ac signal for transmission is called a "modulator." The device that converts the ac tone back into a properly shaped digital signal is called a "demodulator." The generally used arrangement, called a "modem" (MOdulator/DEModulator), combines both functions. The first of the present day type modems was developed in the 1950's primarily by MIT's Lincoln Laboratories for transmitting digital signals between air defense sites over analog telephone lines. The efficacy of these devices has improved steadily since then by improvements from a number of organizations. Today we are reaching perhaps 50% of the theoretical transmission rate of the conventional telephone line. Modem cost has been declining steadily as the signal processing within the modem is itself converted to digital processing. This is performed effectively using large scale integration semiconductor chip circuits. The telephone plant in existence today provides a widespread low entry cost access channel for many of today's computer applications. Bandwidth Availability The figure of merit for a modem is the number of bits per second of digital information that can be transmitted over a conventional telephone channel. Data rates on the order of 300 bits per second (about 30 characters/second) simultaneously in both directions, are readily achieved at low cost without electrical connection using acoustic coupling over any telephone line. Data rates as high as 9600 bits (one way)(about 960 characters per second) are achievable over most (but not all) voice grade circuits with relatively expensive state-of- the-art technology. Transmission carriers lease full-time circuits with greater bandwidths than voice circuits. Even millions of bits per second can be transmitted over television bandwidth channels. Naturally, the tariffs for such arrangements are expensive, and the applications for such channels limited. Multiplexing While broad bandwidths are technically feasible, most applications can be satisfied by low data rates. A local telephone call or circuit rental has historically been inexpensive, while long distance circuits have been expensive. The cost of a short period terminal connection to a computer on the other side of the country -- an expensive business -- is equal to a long distance voice call, albeit few bits may be exchanged. The answer, of course, is combining many users' signals to share a single expensive long distance channel. Many telegraph, and later, data signals were combined together on a single line. This is called multiplexing. One form is called "frequency multiplexing." Here each user is assigned a different frequency and the individual frequencies are separated at the receiving end by tuned filters. Alternatively, time division multiplexing uses each user's bit streams interleafed sequentially. Fundamental to the notion of multiplexing is the fact that each data user will transmit nothing most of the time. The fastest typist at a keyboard generates only a few tens of bits per second. For the hunt-and-pecker, it is only a few bits per second. The telephone channel can readily transmit 4800 bits per second providing a potential for sharing. Statistical Multiplexing When there is a cluster of data users who wish to send signals to a distant computer, a single expensive circuit can be used. To pack more users onto the line means taking advantage of statistic that assumes that, at any one instant of time, only a portion of the possible users would want to transmit. And, to make the averaging even more effective, advantage can be taken of readily available digital storage to hold data to smooth out the peaks. "Time buffering" increases the efficiency of the multiplexing process. So devices were developed that combined signals from many users and elegantly packed them together for transmission over a single line. Spacial Multiplexing - Packet Switching Multiplexing is highly effective if all users who wanted to intercommunicate with one another are in only two sites. If the potential users are spread all over the map, the same statistics can be utilized. Here it is necessary to identify uniquely where the bits came from, where they are to go, and add a little housekeeping data to detect damaged packets in transit. To minimize the processing equipment the bits are formed into consistently organized packages called packets. Such packet switching appears on the ascendancy. Most widespread computer communications networks to be built in the future probably will use packet switching, at least in part. Packet switching networks are now being built in the U.S., Canada, Europe and Japan. Interconnection between these networks is planned. The advantages of packet switching include: 1) A very robust structure can be built that readily permits building systems whose operation is much more reliable than its elements (communications lines and packet switching centers). 2) It provides the highest degree of statistical averaging to make most effective use of the basic resource. 3) The standard format package simplifies full effective interconnection between completely different computer systems and terminals. 4) No better alternative is in sight for most user- to- computer and most computer network applications. Future Directions for Digital Networks Packet switching, while important, is by no means the end of the development for future computer communications systems. The telephone plant itself is moving on to a mostly digital structure for a number of economic advantages. In the process it is looking more like a computer communications network but with tremendously greater data handling capacity. One might argue that the two really the same. But, it is like saying that boys are just like girls. The minor differences are of major significance. An argument can even be raised that the conventional telephone set is nothing more than a "terminal" which generates and accepts commands and which receives and generates data to other such instruments, with the "data" being digitalized voices. A single digital voice channel in today's digital telephone systems carries 64000 bits/second without a conventional modem. Clearly here is a tremendous capacity. There are future needs that could well use some of these high capacity trunks of the telephone plant. There is a need in the distributed processing area for large bandwidths. In the case of inter-computer file transfers or access, bandwidths on the order of two to three megabits are appropriate while for the interchange of high fidelity pictorial information, ten to fifty megabits are necessary. Broad bandwidths are not needed everywhere. A duality exists in the local distribution of data that currently exists in the local vs toll distribution of voice; namely the cost of local circuits is much less than those of toll. In the case of local distribution of high speed data, new systems such as the Pierce loop, the UC Irvine Ring, and the Xerox Ethernet have been proposed and constructed to attack the problem. These systems are marginally applicable to the distribution of such capabilities over large distances. The interfacing of such local data networks to national and international networks represents one of the present edges of computer communications technology. ------- 10-MAY-78 22:13:42-PDT,323;000000000001 Mail from OFFICE-1 rcvd at 18-FEB-77 0459-PST Date: 18 FEB 1977 0448-PST Sender: DAMES at OFFICE-1 Subject: MSGGROUP# 463 Request for MSGGROUP #462 From: DAMES at OFFICE-1 To: MSGGROUP at ISI Cc: DAMES Message-ID: <[OFFICE-1]18-FEB-77 04:48:54.DAMES> REQUEST COPY OF SUBJECT MESSAGE. TOM DAMES ------- 10-MAY-78 22:13:42-PDT,258;000000000001 Mail from CCA-TENEX rcvd at 18-FEB-77 0859-PST Date: 18 FEB 1977 1158-EST From: TOM at CCA Subject: MSGGROUP# 464 MSGGROUP # 456 CB COMPUTER MAIL DRAFT To: MSGGROUP at ISI cc: TOM I WOULD BE HAPPY TO RECEIVE IT BY SNDMSG. --TOM MARILL ------- 10-MAY-78 22:13:42-PDT,489;000000000001 Mail from OFFICE-1 rcvd at 18-FEB-77 1022-PST Date: 18 FEB 1977 1004-PST From: PANKO at OFFICE-1 Subject: MSGGROUP# 465 NASA, the USPS & CMS To: [ISI]Mailing.List: cc: rulifson at PARC-MAXC, taylor at PARC-MAXC, cc: schlesinger at PARC-MAXC, metcalfe at PARC-MAXC According to a reliable source, the director of NASA will meet with Jimmy Carter tomorrow and suggest that NASA develop an electronic mail system for the U.S. Postal Service. No comment. ------- 10-MAY-78 22:13:42-PDT,828;000000000001 Mail from CMU-10A rcvd at 19-FEB-77 1701-PST Date: 19 Feb 1977 1957-EST Sender: PHILIP.KARLTON at CMU-10A Subject: MSGGROUP# 466 Karlton's Introduction From: PHILIP KARLTON(N810PK01) at CMU-10A To: msggroup at ISI - - - - Greetings, I am an implementor (RDMAIL for the CMUx PDP-10 systems). My first name is Phil. Due to last name conflicts, to get mail to me specifically it is necessary to address it to "PHILIP KARLTON at CMU-10A" (or "PHILIP.KARLTON at CMU-10A" if your mail system will not handle spaces in the middle of user names). I have been told that I have no time to do maintainence on RDMAIL but from time to time I get a chance to fix a few bugs. I am currently on the staff here at CMU but will (in about 10 months) become a graduate student once again. Phil ------- 10-MAY-78 22:13:43-PDT,740;000000000001 Mail from SRI-AI rcvd at 22-FEB-77 0205-PST Date: 22 Feb 1977 0203-PST From: Geoff at SRI-AI Subject: MSGGROUP# 467 Self Introduction: GEOF@SRI-AI To: MsgGroup at ISI I am Geoff Goodfellow, The ARPANET Liaison, User Services department, Tenex-Systems Wizard, Packet Radio/Internetiing hacker & .Other at the Artificial Intelligence Center at the Stanford research Inst, here in Sunny Menlo Park, Calif. I am interested in the future of Electronic Mail, and the uses of computers as a means of communications between people, and how it is likely that this is surely going to be something everyone will hopefully be able to utilize in the times to come. I live at GEOFF@SRI most of the time. [Geoff] ------- 10-MAY-78 22:13:43-PDT,1555;000000000001 Mail from OFFICE-1 rcvd at 23-FEB-77 1646-PST Date: 23 FEB 1977 1643-PST From: PANKO at OFFICE-1 Subject: MSGGROUP# 468 Introduction: Ra3y Panko To: msggroup at USC-ISI cc: panko My name is Raymond R. (Ra3y) Panko. The 3 appeared in a computer accident many years ago at Boeing. Although my undergraduate degree was in physics, I am primarily a social scientist. I have an M.B.A. with emphasis in marketing research and organizational communication. My Ph.D. is in communication (as a social science discipline). I am currently a communication analyst at Stanford Research Institute, in the Telecommunications Sciences Center. My current research areas are 1) market studies of computer mail, 2) studies of organizational communication (especially office communication), and 3) office automation systems. I have recently completed the preparation of a bibliography on office communication, in conjunction with Roger Pye of Communication Studies and Planning. I am currently trying to determine what general functions computer mail systems will offer in the 1980s. I have written two papers of potential interest to the group. The first is "The Outlook for Computer Mail," to be published in Telecommunications Policy. The other paper, still in preparation, is a speculative paper on CB Computer Mail -- the possible mating of computer mail with hobby computing. Ra3y Panko Stanford Research Institute 333 Ravenswood Ave. Menlo Park, CA 94025 (415) 326-6200, ext. 4213. Panko at Office-1 ------- 10-MAY-78 22:13:43-PDT,471;000000000001 Mail from PARC-MAXC rcvd at 24-FEB-77 0102-PST Date: 24 FEB 1977 0100-PST From: METCALFE at PARC-MAXC Subject: MSGGROUP# 469 Re: Request for Voluntary Introductions To: STEFFERUD at USC-ISI In response to your message sent 16 FEB 1977 2217-PST I am Bob Metcalfe at Xerox's Systems Development Division in Palo Alto (3333 Coyote Hill Road, 94304) 415-494-4516. We are interested in electronic mail as it might be included in office systems. /Bob ------- 10-MAY-78 22:13:43-PDT,460;000000000001 Mail from BBN-TENEX rcvd at 24-FEB-77 1013-PST Date: 24 Feb 1977 1311-EST From: PERLINGIERO at BBN-TENEX Subject: MSGGROUP# 470 Request for CB Computer Mail Draft To: STEFFERUD at USC-ISI cc: PERLINGIERO In response to your message sent 16 FEB 1977 1942-PST Stef: Would you please either send me copies of msgs. nos. 456 and 462 or tell me the name and password I need to FTP them myself? Thank you, /Clara Perlingiero@BBN ------- 10-MAY-78 22:13:43-PDT,1330;000000000001 Mail from OFFICE-1 rcvd at 7-MAR-77 1745-PST Date: 7 MAR 1977 1724-PST From: PANKO at OFFICE-1 Subject: MSGGROUP# 471 Announcement of Move from SRI To: [ISI]Mailing.List: cc: rulifson at PARC-MAXC, taylor at PARC-MAXC, cc: sutherland at PARC-MAXC, metcalfe at PARC-MAXC < PANKO, ANNOUNCEMENT.NLS;1, >, 25-FEB-77 14:03 RA3Y ;;;; In the Autumn, I will probably be leaving Stanford Research Institute, to take a faculty position at the University of Hawaii. Although all arangements have not been completed, the move is fairly certain. I will be in the Business School, with a joint appointment in organizational behavior and marketing research. I hope to start a program on office automation. Until I leave SRI, I will continue to conduct contract research here. I will honor all research committments, and I hope to initiate and complete some new research projects before I leave. My relationship with SRI after I leave has not been decided yet. I will continue to do research at the U of H. There is even some chance that the school's location in the Pacific basin will create new research opportunities. In general, I hope to focus on the impact of computer mail and office automation on the organizational collaboration process. Aloha, Ra3y Panko ------- 10-MAY-78 22:13:43-PDT,1019;000000000001 Mail from OFFICE-1 rcvd at 12-MAR-77 0318-PST Date: 11 MAR 1977 1616-PST From: PANKO at OFFICE-1 Subject: MSGGROUP# 472 NTC'77 Computer Mail Session To: [ISI]Mailing.List: cc: rulifson at PARC-MAXC, taylor at PARC-MAXC, cc: sutherland at PARC-MAXC, metcalfe at PARC-MAXC I will be chairing a session on computer mail this December 4, at NTC'77 in Los Angeles. The session will draw together designers from ARPANET Computer Mail, computer teleconferencing, commercial mailbox services, the teletypewriter network world, and, perhaps, facsimile. I have long held the feeling that designers from different forms of computer-based human communications need to get together to exchange ideas. Perhaps this forum will stimulate more cross-disciplinary discussions. I hope many of you can make it there to listen to the papers and take part in a rump session I am planning for after the formal session. More to come in a few weeks. Ra3y Panko Stanford Research Institute ------- 10-MAY-78 22:13:43-PDT,1101;000000000001 Mail from USC-ISI rcvd at 14-MAR-77 1817-PST Date: 14 MAR 1977 1806-PST Sender: FARBER at USC-ISI Subject: MSGGROUP# 473 Here we go!! From: FARBER at USC-ISI To: [ISI]Mailing.List: Message-ID: <[USC-ISI]14-MAR-77 18:06:35.FARBER> Quoting from the April 1977 edition of Radio-Electronics page 93 "....One item that's definitely a computer and not a game is the $495. Personal Electronic Transaction (PET) device being readied for sale by Commodore, developed by its subsidiary MOS Technology. With its built-in cathode-ray tube readout, the unit has an alphanumeric keyboard and is designed as an all-purpose household computer. It will accept specially programmed audio cassettes, or the owner can program his own. Since it speaks standard computer language, it can communicate by phone with other computers or central information banks." Other sources speculate summer for sales and one suspects the programming is BASIC. Remember that the sales of the HP 35 at $395 1974-1975 dollars was unbelievably large. ------- 10-MAY-78 22:13:43-PDT,20189;000000000001 Mail from OFFICE-1 rcvd at 18-MAR-77 1608-PST Date: 18 MAR 1977 1608-PST Sender: PANKO at OFFICE-1 Subject: MSGGROUP# 474 Draft Survey of ARPANET Computer Mail From: PANKO at OFFICE-1 To: STEFFERUD@USC-ISI Message-ID: <[OFFICE-1]18-MAR-77 16:08:09-PST.PANKO> B. COMPUTER-BASED MEDIA Compared to other forms of electronic mail, there is a strong kinship between ARPANET computer mail, computer teleconferencing, and commercial mailbox services on time-sharing networks. What the three forms have in common is a single ancestor; all can trace their heritage to the "mailbox" services that have been available on even the earliest time-sharing systems. Mailbox programs allow one user to send brief messages to other users or to operators. These messages either go directly to the terminal of the recipient or are delivered the next time the receiver logs into the system. Yet actual communication between designers of different media, although growing, is still very small. ARPANET designers have been mostly artisans, who have seldom reported on their systems in the open literature. Conferencing designers have tended to be social scientists, who have prepared voluminous reports, usually dealing with impacts on people. Commerical mailbox designers have business people, who have seldom even communicated with one another. Before time-sharing, users had to hand-carry programs to the computer center. While this cumbersome process had numerous drawbacks, it did bring progammers into contact with their colleagues and with computer operators. If they had questions, they could walk a few feet and ask them. But time-sharing systems separated users from one another by hundreds of yards, so getting help became difficult. Mailbox programs were installed to ease communication problems. The very first time-shared computer, CTSS at MIT's Project MAC, offered a mailbox program (*), and few time-shared systems since then have not followed suit. * Crisman, P.A., ed., The Compatible Time-Sharing System, A Programmer's Guide (2nd Ed.), The MIT Press, Cambridge, Massachusetts, 1965, Section AH.9.05, quoted in Stuart L. Mathison and Philip M. Walker, Computers and Telecommunications: Issues in Public Policy, Prentice-Hall, Inc., Englewood Cliffs, New Jersey, 1970. A related facility is "linking," in which two terminals are tied together so that each user can see what the other is typing. If mailbox delivery resembles postal service or interoffice mail, linking resembles conversational interactions, such as telephone calls or face-to-face meetings. While linking and mailbox services can be quite distinct from one another on any given system, they really form a conceptual continuum of asynchronous and synchronous service. 1. ARPANET COMPUTER MAIL During the 1960's, a substantial amount of the world's advanced computer research was funded by the Advanced Research Projects Agency (ARPA) of the U. S. Defense Department. To make ARPA-funded software more generally available, ARPA began funding the development of a national packet-switched computer network, later known as the ARPANET. The first lines and switching computers were installed in 1969, but it was not until the network's file transfer protocols (FTPs) were refined, in 1972, that the network became fully operational. In 1975, operation of the network was transferred to the Defense Communications Agency. Today, there are 182 host computers on the network. Half of these are large computers, the other half minicomputers. The most common large computer on the network is Digital Equipment Corporation's PDP-10. Most of the 36 PDP-10s on the network use the TENEX operating system, developed by Bolt, Beranek and Newman, under ARPA funding, in the early 1970's. In 1972, just before the FTP was established, Ray Tomlinson of BBN developed message sending and reading programs for TENEX. The sending program was called SNDMSG, the reading program READMAIL. Originally, SNDMSG and READMAIL were written to handle mail flows within individual PDP-10s. Late in 1972, the package was rewritten to handle message distribution over the network, via the FTP. The subsequent development of computer mail is difficult to characterize chronologically. There were several streams of development, in which successive programs refined earlier efforts. But each stream borrowed extensively from the others, adopting attractive innovations developed in other ARPANET systems. The oldest stream grew out of refinements to READMAIL. Although SNDMSG was gradually refined, primarily by Julie Sussman, its evolution was gradual and limited. Just the opposite was true for the message reading side of the package. In 1973, Larry Roberts at ARPA wrote RD. This program was coded in TECO (a text editing program) macros. It offerred several new features, as did its successor NRD, which was written later in 1973, in the SAIL language, by Barry Wessler at Telenet. In 1974, Martin Yonke and John Vittal at U.S.C.'s Information Sciences Institute (ISI) wrote WRD in 1974. Later that year, Yonke wrote BANANARD, and, in 1975, Vittal wrote MSG. MSG is currently the most popular message-reading program on the ARPANET. These programs, from READMAIL through MSG, were written for TENEX PDP-10s. Also written for TENEX machines were two other programs, developed more or less simultaneously in the middle of the decade at Bolt, Beranek and Newman. HG (the checmial symbol for mercury) was written by James Calvin in 1974. Another program, MAILSYS, was developed under Ted Myer in 1974. MAILSYS, which was also called XMAIL, was the larger project. As discussed below, it was later expanded to become HERMES (TM), for the Military Message-Handling Experiment. MAILSYS, and to some extent HG, combined both mail reading and mail composition functions. In contrast, the READMAIL-MSG stream either had no message composition tools or transferred the user to SNDMSG for composition. In 1975, the Dynamic Modeling System project at MIT developed a message program called MSGDMS. This program was developed primarily by Mike Brooz, working under Al Vezza. MSGDMS's major innovation was its inverted index system, which allowed for very rapid retrival of old messages. Another major innovation was the sophisticated use of "background" processes, which periodically run housekeeping chores during unused computer cycles. In addition, MSGDMS was linked to the ARPANET Data Computer, a terrabit storage center. This allowed old messages to be archived, thus reducing online storage costs (which can be heavy for frequent computer mail users) yet still allowed reasonably fast retrieval of archived messages. MSGDMS was originally written under the ITS operating system developed at MIT. It was later modified to run under TENEX, under the name XMAIL. The fourth major stream of development began in 1973, when ISI released its report "Consolidation of Telecommunications on Oahu" (COTCO). The report, based on an extensive study of naval communications on Oahu, recommended the the application of computer mail to operational military environments. ARPA then funded the Information Automation project at ISI, under Rob Stotz, to develop a terminal and computer mail software for a military environment. In 1975, ARPA expanded the effort and funded the development of competitive computer mail programs at MIT and BBN. The ISI program is called SIGMA. The BBN program, HERMES, was based on MAILSYS. The MIT program is as yet unnamed. All three programs are written to run on the Hewlett-Packard 2645A terminal, under operating system software developed by ISI. This year, one of the three programs will be tested extensively in the Military Message-Handling Experiment in Oahu. The other two programs will be tested, but on a more limited basis. MITRE is currently evaluating the three systems. During the experiment on Oahu, MITRE will conduct an evaluation of the results. Our discussion has passed over many computer mail programs whose application has been more limited but which introduced many interesting and useful features. One current program that bears special mention is MS, being developed by Dave Crocker and Bill Crosby, under Bob Anderson, in RAND's Personal Computing project. MS (pronounced "Miz") runs on a RAND-UNIX PDP-11 minicomputer. The UNIX operating system, which was developed at AT&T, has sophisticated text-editing features. ARPANET computer mail is almost bewildering for its diversity. Some programs were developed under intense direct funding. Others were written in programmers spare time. Despite this diversity, ARPA has been able to coordinate network mail development, albeit loosely. Minimal mail header standards for FTP have been created under ARPANET Requests for Comment 680 and 720. In addition, the ARPANET has a working committee on Computers and Human Communication (CAHCOM), which is chaired by Dave Farber at Irvine. There is also a loosely-knit Message Service Group (MSGGROUP), which has about 60 members and has been conducting an active general dialog over the network since mid-1975. Although much standardization still needs to be done, messages can already be sent among most TENEX and non-TENEX hosts on the ARPANET. Perhaps the most useful service provided by the network to mail users is the ARPANET Directory, which is now funded by the Defense Communications Agency. The Directory, published by the Network Information Center (NIC) at SRI, is like a telephone book -- giving the names of ARPANET mail users, the host computers to which their mail is delivered, their postal addresses, their telephone numbers, their unique network idents (which are like telephone numbers, license plates, or TWX acronyms), and the names of network groups to which they belong. A discussion of network communication would be incomplete without a discussion of "linking," in which two distant terminals are locked together. When two users link terminals, each can see what the other types. They can type messages back and forth in a completely conversational style. They can even execute a program together and discuss the results. Most large ARPANET hosts have some form of linking for two parties using the same host. In addition, through the RSEXEC system developed at BBN, a user on any TENEX computer can locate and link to any other loged-in of a TENEX machine on the ARPANET, and with users and several other types of machines as well. RSEXEC is essentially a distributed multi-host operating system with many interesting characteristics. RSEXEC linking across hosts has been available since 1972. A number of non-TENEX hosts now offer RSEXEC. In NLS, a system developed under Engelbart at Stanford Research Institute, shared-screen teleconferencing allows two users to link video displays or to link a video display to a wall screen projector. Shared-screen teleconferencing in conjunction with a telephone call, has been used to train distant NLS users. Shared screen teleconferencing has also been used to augment face-to-face meetings. Normally only two parties are linked at one time. Multi-party linking would require considerably more discipline than standard linking tools provide. Some progress has been made in providing multi-user, multi-host linking. The most notable example is TALK, developed by Jim Calvin at BBN. TALK allows group conferencing among users on several different hosts. The ARPANET has been used on several occasions to handle true computer teleconferencing systems. Until late 1974, for example, the FORUM teleconferencing system, developed at the Institute for the Future under Vallee, Amara, Lipinski, Miller, and Helmer, used the ARPANET for its experiments. But FORUM has not played any long-term role on the NETWORK, except for some limited continuing uses at ISI. Composition requires editing, and virtually all message-sending systems provide some editing tools. Editing can become quite extensive, including right-margin justification, automated spelling-correction, and customized formatting. In advanced systems, the dividing line between computer mail and word processing is becoming quite blurred. We have held until the last our discussion of the most conceptually sophisticated computer mail system on the ARPANET: NLS Journal Mail. NLS, as noted above, was developed at SRI. Development began in 1963, primarily under ARPA funding. It continues today, under mixed funding. Overall, NLS is an integrated office automation system, offering extensive document composition tools, forms systems, and other office-related tools. In 1970 and 1971, SRI developed the Journal Mail subsystem, to distribute messsages, pre-prepared documents, data, line-drawn pictures, and other information. Because NLS was developed in an environment where long documents were common, it developed facilities for delivering long documents without inconveniencing users -- a facility no other ARPANET computer mail system provides. Journal Mail was built as a working tool for a complex programming effort. As a result, careful thought was given, in 1969, to the problems raised by personnel turnover, limits on human recall, general communication processes on complex projects and other facts of life that require formal dialog in organizations. The Journal design embodied many correspondence control mechanisms, although most of these were not implemented fully. The original design called for directory assistance-type functions and dialog recording in a multi-host environment, and many design innovations were made on this area. The original design also called for logical entities called "sets" of messages, which could be treated as personal message files, as teleconference transcript files, as successive versions of a controlled document, and so on. Potentially, sets could provide a basic logical architecture for complex communication processes. Many concepts in the original journal design were not implemented, but others were. Features that were developed, such as an ability to handle long documents and on-line "directory assistance" functions make journal mail quite powerful. In addition, NLS Journal Mail has "hooks" into other media. It can accept files prepared off-line on several terminals; it can deliver mail to a non-NLS user via standard ARPANET computer mail; and, if a user is not a network user, it prints the item for postal distribution. A number of ARPANET mail programs are beginning to be used by nonresearchers. NLS Journal Mail, HERMES, and MSG are being used by operational military organizations with access to the ARPANET. The U.S. Army Material Development and Readiness Command (DARCOM) for example, has over 200 computer mail users. Both NLS Journal Mail and HERMES have non-ARPANET users. HERMES service is available via Telenet, a commercial computer netwwork owned in part by BBN. NLS service, including Journal Mail, is sold f.o.b. Cupertino. While ARPANET computer mail is the most visible communication-oriented activity on the network, there are scattered ancillary developments that indicate new ways to augment the usefulness of computer mail. We consider just three here: calendar, bulletin board, and forms systems. As discussed in Working Paper II, office workers communicate most frequently with people who are "close" in the organizational sense, i.e., the average individual has far more communication with his or her office mates than with people in other divisions in thecompany, still less with people in other companies. It may follow also be true that a person's most intense form of communication is with himself or herself in the form of notes and reminders. At least one good appointment system has been built on the ARPANET. This is CALENDAR, written at BBN by Ted Strollo. CALENDAR is not a sophisticated system, nor is it stylistically easy to use, but it is quite workable and points to new directions for development. We have no statistics on its daily use by a typical user, but we suspect it rivals or surpasses the frequency of mail system use. Moving up to the level of the working group, Richard Kahler at Stanford Hospital's SUMEX-AIM project has written a bulletin board program, called BBD, that runs under TENEX. BBD is written in SAIL, and it is a well-designed, well-conceptualized system. It is very similar in its command style to ARPANET computer mail systems, especially MSG; in fact, it is tied to MSG, although rather loosely. Strictly speaking, BBD should not be listed under ARPANET computer mail, since it is not an ARPA-funded activity, but it is closely tied to the ARPANET community. Conceptually, it would be possible to build calendar and bulletin board systems as integral parts of computer mail systems. Conceptually, both appointments and bulletin board entries can be viewed as messages, albeit with some special characteristics such as expiration dates and notification cycles. With careful design, it would be possible for users to read their bulletin boards with his or her normal mail reading commands, to send notices to their calendar via normal computer mail, to receive appointment reminders via computer mail, and to read through filed messages as easily as messages in the user's standard "in box" (this last feature is already common). A mail system well-tied to calendar and bulletin board systems would probably receive a much higher level of use than a stand-alone mail system. There has only been sporadic development in forms systems, although there are a few examples of special-purpose systems designed to handle one or two specific types of forms. One interesting experimental system system is NEWFORMS, a subsystem of NLS. In addition, the three systtems designed for the Military Message Handling Experiment can handle form composition quite easily. The HERMES command "Compose," for example, can be extended to include the name of a form, e.g., "Compose Invoice." Most existing forms systems, unfortunately, do little more than prompt the composer to input various fields of information. But a forms system should also be able to collect a good deal of background information itself, so that it can fill in various fields automatically, without requiring human assistance. The NLS forms system has some capabilities in this area, but they are limited. Future forms systems will amost certainly search data bases and do complex data manipulations in order to fill out fields. Recently, Michael D. Zisman has suggested that the challenge in forms processing is not task assistance but task recognition (*). According to Zisman, a system should not only assist the user in filling in forms, but it should automate the forms process by sensing when a form should be sent out for completion. As part of his dissertation at the Wharton School of the University of Pennsylvania, Zisman is developing a task-recognition system to automate the process of reviewing and publishing journal articles. The main activity of Zisman's system is to send out "tickler" messages when certain events occur, then checking on the status of responses to these messages as time goes on. (*) Presentation at Stanford Research Institute, March 6, 1977. ------- 10-MAY-78 22:13:44-PDT,529;000000000001 Mail from OFFICE-1 rcvd at 22-MAR-77 0649-PST Date: 22 MAR 1977 0638-PST From: PANKO at OFFICE-1 Subject: MSGGROUP# 475 RE: Draft History of ARPANET Computer Mail To: [ISI]Mailing.List: cc: rulifson at PARC-MAXC I have just finished the draft of a history of ARPANET computer mail. I would appreciate your looking it over for errors and missed points. It is item #474 in the MSGGROUP dialog at ISI. It is 20189 characters long. Ask Stef for a copy if you would like to see it. Ra3y Panko ------- 10-MAY-78 22:13:44-PDT,1716;000000000001 Mail from USC-ISI rcvd at 23-MAR-77 0036-PST Date: 23 MAR 1977 0018-PST Sender: STEFFERUD at USC-ISI Subject: MSGGROUP# 476 [Communications policy seminar at MIT] From: STEFFERUD at USC-ISI To: [ISI]Mailing.List: Message-ID: <[USC-ISI]23-MAR-77 00:18:23.STEFFERUD> Begin forwarded message -------------------- Date: 22 Mar 1977 at 2224-PST From: Greep at Rand-Unix To: Stefferud at Isi, Farber at Rand-Unix, Dcrocker at Rand-Unix Cc: Greep at Rand-Unix Subject: Communications policy seminar at MIT Message-ID: <[Rand-Unix]22-Mar-77 22:24:31.Greep> Mail from RAND-UNIX rcvd at 22-MAR-77 2225-PST MAIL 1: LRH@MIT-MC 03/22/77 19:20:51 Re: Policy Seminar M.I.T. Communications Policy Seminar "Innovation Strategies in the Electronic Mail Marketplace" Speaker: Dr. Marvin Sirbu, Research Associate, Center for Policy Alternatives, MIT Respondant: Dr. John Rockart, Director, Center for Information Systems Research, MIT Thursday, March 24th 4 - 6 p.m. E52-143 Several new communications services are vying for the emerging electronic mail market, including remote facsimile and interconnected word processing. Market shares among these services will be affected by the office innovation strategies pursued by equipment users and suppliers ;by the service and pricing strategies of the communications carriers; and by policies set by the Congress, the FCC, the Postal Service and other agencies. Dr. Sirbu will assess several possible innovation strategies in the context of these market and regulatory developments. -------------------- End forwarded message ------- 10-MAY-78 22:13:44-PDT,881;000000000001 Mail from USC-ISI rcvd at 23-MAR-77 1627-PST Date: 23 MAR 1977 1620-PST Sender: STEFFERUD at USC-ISI Subject: MSGGROUP# 477 Re: [Communications policy seminar at MIT] From: STEFFERUD at USC-ISI To: Burchfiel at BBNA, Henderson at BBNA, Mooers at BBNA, To: Myer at BBNA, Vittal at BBNA, Heitmeyer at BBN, To: Mathison at BBN, Perlingiero at BBN, Walker at BBN, To: JZS at CCA, Tom at CCA, RCT at CCA, Broos, KLH at MIT-AI, To: RMS at MIT-AI, MSGGRP at MIT-DMS, Vezza at MIT-DMS, To: Frankston at MIT-MULTICS, Pogran at MIT-MULTICS, To: CBF at MIT-ML Cc: Stefferud, MsgGroup Message-ID: <[USC-ISI]23-MAR-77 16:20:17.STEFFERUD> In-Reply-To: <[USC-ISI]23-MAR-77 00:18:23.STEFFERUD> Hey - How about some one of you guys who are close to MIT, and who attend the announced seminar writing a brief report on it for MsgGroup? Thanks, Stef ------- 10-MAY-78 22:13:44-PDT,4156;000000000001 Mail from USC-ISI rcvd at 25-MAR-77 1014-PST Date: 25 MAR 1977 0944-PST Sender: STEFFERUD at USC-ISI Subject: MSGGROUP# 478 [Pogran AT MULTICS: Report on Sirbu Seminar] From: STEFFERUD at USC-ISI To: [ISI]Mailing.List: Message-ID: <[USC-ISI]25-MAR-77 09:44:52.STEFFERUD> Begin forwarded message -------------------- Date: 03/25/77 1209-est From: Pogran at MIT-Multics To: Stefferud at USC-ISI Subject: Report on Sirbu Seminar Mail from MIT-MULTICS rcvd at 25-MAR-77 0908-PST Sirbu Seminar To: Stefferud at USC-ISI A brief report on the seminar "Innovation Strategies in the Electronic Mail Marketplace" given by Dr. Marvin Sirbu of the MIT Center for Policy Alternatives at MIT, March 24, 1977. Sirbu began with an objective survey of the state of the art of various sorts of electronic mail systems: communicating word processing systems, facsimile systems, and computer-based message systems. He presented statistics on market size, growth patterns, etc. Since computer-based message systems were less familiar to most of the audience than were the other two, he went into somewhat more detail discussing them, covering (by brief description, not in-depth discussion) such aspects as message preparation, transmission, storage and retrieval, and getting into such current issues as authentication and privacy. He also discussed the regulatory aspects, including the December FCC decision regarding resale of data communication facilities. Sirbu quoted Ra3y Panko's SRI study with regard to costs. He then got into future developments, discussing the Postal Service's mail volume pie, and what share of the various pieces could be converted over to the various electronic mail systems. He discussed possible marketing strategies for word-processing and facsimile equipment. He emphasized that the systems (including computer-based message systems) must become inexpensive enough and simple enough to operate, so that they could become "ubiquitous" in the well-equipped office of the future. At this point, I had high hopes that the discussion following the presentation might get into the computer-based message system area, with members of the audience asking for more details about them, etc. However, the format of the seminar series called for respondents, and the first respondent, Howard Anderson, of the Yankee Group, a Cambridge-based marketing research firm which is involved in market surveys in the electronic mail field, dashed my hopes totally. Anderson launched into a boisterous discussion of the facsimile field -- who was buying, the treatment of fax as an aid to the individual office rather than a communication function within the purview of the corporation's communications officer, the impact of SBS's dish-on-the-roof as the next decade's corporate status symbol, and as a source of a cheap high-bandwidth connection, etc. He dismissed computer-based message systems out-of-hand as being too complex for the ordinary office worker, and left me, at least, with the impression that he considered them to be playthings rather than of practical market value. In short, Anderson promoted fax, fax, fax, and, essentially, ensured that the discussion which followed the second respondent's talk would be paralyzed around facsimile. The second respondent was Dr. John Rockart, the director of the MIT Center for Information Systems Research, which is part of the Sloan School of Management. Rockart delivered a low-key pitch for observing how IBM goes in the electronic mail field, for as IBM goes, so goes the marketplace. As I indicated above, the discussion which followed was paralyzed around the facsimile field, and nothing further was said about computer-based message systems. MsgGroupers in attendance included Ra3y Panko, Al Vezza, Bob and Charles Frankston, Lyman Hazelton, and myself. My apologies to anyone I've left out. Ken Pogran, Scribe -------------------- End forwarded message ------- 10-MAY-78 22:13:44-PDT,596;000000000001 Mail from USC-ISI rcvd at 26-MAR-77 1129-PST Date: 26 MAR 1977 1120-PST Sender: STEFFERUD at USC-ISI Subject: MSGGROUP# 479 Add BZM@CMUA, Walsh@OFFICE-1, Pine@SRI-AI From: MSGGROUP at USC-ISI To: [ISI]Mailing.List: Cc: BZM at CMUA, Walsh at OFFICE-1, Pine at SRI-AI Message-ID: <[USC-ISI]26-MAR-77 11:20:40.STEFFERUD> WELCOME TO MSGGROUP - IF YOU WISH TO SEND A BRIEF INTRODUCTION OF YOURSELF, WE WILL FILE IN IT IN [ISI]INTRODUCTIONS.MSG FOR THE BENEFIT OF THOSE WHO WISH TO KNOW WHO IS IN HERE. JUST SEND ONE COPY TO MSGGROUP@ISI - ENJOY, STEF ------- 10-MAY-78 22:13:45-PDT,514;000000000001 Mail from CMU-10A rcvd at 26-MAR-77 1916-PST Date: 26 Mar 1977 2217-EST From: BZM at CMU-10A Subject: MSGGROUP# 480 Introduction for Bruce Nelson To: MsgGroup@ISI Sender: BRUCE.NELSON at CMU-10A Message-ID: [CMU-10A] 26 Mar 1977 22:17:51 Bruce Nelson I'm Bruce Nelson (alias BZM, and the curious can seek enlightenment), a graduate student in the CS Department at CMU. I'm primarily a "critic", but also a potential implementor (of the Secure Mail System for C.mmp-Hydra). ------- 10-MAY-78 22:13:45-PDT,491;000000000001 Mail from USC-ISI rcvd at 26-MAR-77 2158-PST Date: 26 MAR 1977 2149-PST Sender: STEFFERUD at USC-ISI Subject: MSGGROUP# 481 Adding Hathaway@AMes-67 and Brian.Reid@CMUA From: MSGGROUP at USC-ISI To: [ISI]Mailing.List: Cc: Hathaway at AMES-67, Brian.Reid at CMUA Message-ID: <[USC-ISI]26-MAR-77 21:49:26.STEFFERUD> SNDMSG requests for fresh copies of [isi]mailing.list should be sent to MsgGroup@ISI, or you can FTP a copy directly. Best, Stef ------- 10-MAY-78 22:13:45-PDT,757;000000000001 Mail from BBN-TENEXA rcvd at 27-MAR-77 1358-PST Date: 27 Mar 1977 1653-EST Sender: HENDERSON at BBN-TENEXA Subject: MSGGROUP# 482 Introduction for Austin Henderson From: HENDERSON at BBN-TENEXA To: MSGGROUP at USC-ISI Message-ID: <[BBN-TENEXA]27-Mar-77 16:53:41.HENDERSON> In-Reply-To: <[USC-ISI]26-MAR-77 11:20:40.STEFFERUD> Introduction of D.Austin Henderson, Jr. (=Henderson at BBN) I am an implementer. I had early responsibility for the design and implementation of BBN's Hermes message manipulation program. More recently I have begun looking at more advanced usage of messages: the buz-word is "office automation", and the idea is integration of the message paradigm into a wider range of our daily activities. ------- 10-MAY-78 22:13:45-PDT,714;000000000001 Mail from UCLA-SECURITY rcvd at 29-MAR-77 0935-PST Date: 29 Mar 1977 0930-PST Sender: mark at UCLA-Security Subject: MSGGROUP# 483 Introduction for Mark Kampe From: mark at UCLA-Security To: msggroup at isi Mark Kampe, system's programmer at ucla-security and ucla-ats My paying work is the development of a data secure version of the Unix operating system. My interests in the message group stem from an interest in distributed processing and distributed resources in general. I've been on the network since '70 and have been active in RJE protocol, the Ants evaluation, the development of ELF, Unix development for ARPA users and performance measurements of the network. ---mark--- ------- 10-MAY-78 22:13:45-PDT,7059;000000000001 Mail from BBN-TENEXA rcvd at 30-MAR-77 0713-PST Mail from BBN-TENEXA rcvd at 25-Mar-77 1342-EST Date: 25 Mar 1977 1332-EST Sender: DDEUTSCH at BBN-TENEXA Subject: MSGGROUP# 484 Electronic Mail seminar at MIT From: DDEUTSCH at BBN-TENEXA To: Myer, Dodds, JMiller, Mooers, Henderson, Vittal, Stevens Cc: DDeutsch, Cc: Burchfiel, DDeutsch Message-ID: <[BBN-TENEXA]25-Mar-77 13:32:46.DDEUTSCH> Redistributed-To: msggroup at ISI Redistributed-By: DDEUTSCH Redistributed-Date: 30 Mar 1977 This is an attempt to summarize the proceedings at yesterday's meeting of the MIT Communications Policy Seminar, entitled "Innovation Strategies in the Electronic Mail Marketplace". Feel free to speak with me if you have any questions. Dr. Marvin Sirbu, a research associate at the Center for Policy Alternitives was the main speaker. He broke the problem into five "office functions": 1) Recording or capturing the message 2) Editing or manipulating it 3) Display and reproduction of a message 4) Storage and retrieval of correspondence 5) Communication, or the actual sending He claimed that these functions are becoming more and more integrated, and further stated that this integration was dependent on three technologies. 1) Word processing. Local editing capabilities and storage as on an IBM Selectric. Several commercial systems are on the market, some with the capability to communicate over normal phone lines. Totally character oriented. Storage media used have gone from paper tape to magnetic card or tape, with floppy disc on the horizon. 2) Facsimile. There was a great deal of concentration on this mode of communication at the seminar. The market is dominated by Xerox, which has set de facto standards for encoding. Transmission is slow, as at 96 bits/inch it takes 6 minutes to transmit an 8 1/2 by 11 page. The hardware necessary accounts for only a third of the total cost of such a system, with the actual communiction costs accounting for the other two thirds. Most communication of this type is within a single company (95%). There is a 30%/year growth in sales of facsimile equipment in the US. 3) Computer based electronic mail systems. Of course the ARPANET was mentioned, but it was also interesting to discover that CETA (the airlines' net) has a message system which has as 25% of its traffic acttual correspondence instead of the reservations which it was designed to carry. While the cost of a message on the ARPANET today is estimated at $.80 (computer .10, labor .10, communications .30, terminal .30) it is estimated that in 1985 this cost will have dropped to $.25. This compares quite favorably to the $4.00 which it costs today to compose and send a letter by ordinary office procedures. Dr. Sirbu outlined what he thought were the design criteria for tomorrow's electronic mail system. These were: Ubiquity-It must be convenient and easy to find access to the system. He mentioned the terminal on everyone's desk as an example. Access points must be inexpensive to obtain and maintain or the system will be limited. Simplicity-There must be ease of operation, with little or no training needed to use the system. Anyone in an office should be capable of operation, there should be no need to train one person and then have him or her tied up performing only one dull operation. Flexibility-There should be no restrictions on length of messages, graphics, typefaces, etc. Reliability-There must be backup hardware (computers, etc.) and backups on the stored messages. It is essential that no message "disappear". Integration-There must be a smooth and reliable interface between all modules of the system (editors, filers, forwarders, etc.). Privacy-There must be positive assurance that correspondence is secure. Possible use of encryption techniques. Protection from unwanted "junk mail" by positive identification of sender and ability to screen against unwanted senders. The market for electronic mail systems was appraised. It is unique in that : 1) A mail system does not tend to depreciate in value, but instead becomes more valuable as more and more information is stored in it. 2) The value of a mail system increases as the square of the number of users. Xerox published its specs in order to increase the number of users on a facsimile system, and was rewarded by having the value of its units inflated because there were more people to send to (and receive from) on the system. 3) Systems proliferate within companies before they link them. The marketing strategies for a word processing system, a facsimile system, and a computer-based system were outlined. I omit the first two here. Dr. Sirbu suggests that the first place to market a computer-based mail system is where the potential user has had some exposure to computers, and already has access to terminals. This eliminates need for any computer education , and takes advantage of pre-existing hardware (the terminals). These markets are found in universities and research communities. The phone company, regulatory issues, and the impact on the US Postal Service were briefly discussed. The low quality of phone lines, and the high costs of long distance telephone transmissions has led to the creation of SBS (a wholely owned subsidiary of IBM,Aetna, and ComSat) which is engaged in building a satellite system designed to connect widely distributed companies by telecommunications links. This service should be less expensive than that offered by AT&T, which is seeking to preserve its monopoly in the field. A more familiar concept, that of a packet network, is overcoming most of the noise problems inherent in the current phone system. Packet switching does not work well for facsimile transmission, which demands a more synchronous form of reception. AT&T is slowly losing ground in its fight to protect itself from competition. Courts have ruled that it cannot ban connection of outside equipment to its lines, the Congress is debating removing protections against competition, and recently the courts have allowed a computerized mail system to act as a carrier. About 22% of the Post Office's business is business mail. It was suggested that much of this will be replaced eventually by Mailgram type arrangements, thus speeding delivery and saving money. A far greater impact on the Post Office will be realized by the spread of Electronic Funds Transfer systems, as about 40% of the mail is made of checks, bills,etc. In closing, it was postulated that burst communications (storing up messages, and then transmitting them in one "burst") may be the wave of the future. IBM, which effectively controls the direction of the market, has heavy investments in centralized, specialized systems, and the lack of money needed to use today's technology limit severely the speed at which the field will advance. ------- 10-MAY-78 22:13:45-PDT,447;000000000001 Mail from OFFICE-1 rcvd at 1-APR-77 0913-PST Date: 1 APR 1977 0913-PST From: Panko at Office-1 Subject: MSGGROUP# 485 Article on New Public Data Networks Sender: HOUGH at OFFICE-1 To: [ISI]Mailing.List: Peter Kirstein (of our very own MSGGROUP) has a very good article, "Planned New Public Data Networks," in the September 1976 issue of Computer Networks. This is an excellent summary of current trends. Ra3y ------- 10-MAY-78 22:13:45-PDT,1728;000000000001 Mail from OFFICE-1 rcvd at 1-APR-77 1004-PST Date: 1 APR 1977 0954-PST From: PANKO AT OFFICE-1 Subject: MSGGROUP# 486 Electronic Mail: $1 Billion a Year Sender: HOUGH at OFFICE-1 To: [ISI]Mailing.List: cc: rulifson at PARC-MAXC, taylor at PARC-MAXC, cc: sutherland at PARC-MAXC, day at OFFICE-1, cc: mcbrearty at OFFICE-1, bedford at OFFICE-1 I have been hearing a number of comments, some from MSGGROUPers, that electronic mail will some day be very large, approaching $1 billion a year. My remark is that it already is a $1 billion a year business. Telex and TWX alone are nearly a quarter billion a year, and other services, such as mailgram and Infocom probably raise public teletypewriter revenues to comfortably over a third of a billion dollars each year. There are already more facsimile terminals and transmission than Telex and TWX terminals and transmissions. While fax is a cheaper medium, its revenues must be somewhere between one hundred million and three hundred million annually. Finally, internal Corporate teletypewriter networks, which are usually called "private wire" networks, cost businesses at least $400 billiona year and may actually be closer to a billion dollars a year. These numbers are rough, but under almost any assumptions you make, electronic mail comes to around $1 billion a year. And there are a few other things, like communicating word processing typewriters. There are already a quarter as many communicating typers as Telex and TWX terminals, and there numbers are growing rapidly. Sorry for the long message, but I think this point is important to keep in mind when we discuss computer mail. Aloha. Ra3y ------- 10-MAY-78 22:13:46-PDT,1662;000000000001 Mail from SRI-KA rcvd at 5-APR-77 0150-PST Date: 5 Apr 1977 0151-PST From: Geoff at SRI-KA Subject: MSGGROUP# 487 The 1st West Coast Computer Faire. To: pc at ISD cc: [ISI]Mailing.List: On April 15th though the 17th in San Francisco Civic Auditorium, The First West Coast Computer Faire - A conference & Exposition on Personal & Home Computers is going to be held. Poop: 7,000 to 10,000 People, 100 conference sessions, Publications and Proceedings being planned, 200 Commercial & Homebrew Exhibits & Special Interest Social Centers. Co-Sponsors include Amateur, Professional, & Educational Groups. Being Planned: * Computer Graphics on Home Computers * Computer-Driven & Computer-assisted Music Systems * Speech Synthesis Using Home Computers * Computers & Amateur Radio * Computer Games: Alphanumeric & Graphic. * Personal computers for the Physically handicapped. * Computers & Systems for Small Business * Tutorials for Hardware Novices & Software Novices * Software Design for Personal Computers * Microprogrammable Microprocessors for Hobbyists * Optical Scanning for Inexpensive Program and Data input. * Floppy Disc Systems for Home Computers * Hardware & Software Standards for Personal systems. * Seminars for Club Leaders, Editors, Organizers, etc. * Personal Computers in Edcuation (Associasted with a Univ. of Calif short course). Some guest speakers include: Fredrick Pohl, John Whitney, Henry Tropp, Ted Nelson. Needless to say this oughta be a fine gathering of Professionals and Hackers alike showing off all of their latest toys. [Geoff] ------- 10-MAY-78 22:13:46-PDT,666;000000000001 Mail from SRI-KA rcvd at 8-APR-77 1553-PST Date: 8 Apr 1977 1552-PST From: Geoff at SRI-KA Subject: MSGGROUP# 488 Artical on ARPANET. To: [ISI]Mailing.List: There is an fine artical on the ARPANET entitled: C Reseach at DARPA 3 by Col. Russell himself in the March '77 issue of SIGNAL. It describes Discussion of current problems in C3, Future needs, Packet Switched Computer-Communications, Packet Satelite Experement, Packet Radio, Packet Speech and Interneting, Network and Sysstem Security, Intelligent Terminals & the Evaluation and transfer of C3 Technology. I hope you all can get a copy of it. [Geoff] ------- 10-MAY-78 22:13:46-PDT,615;000000000001 Mail from USC-ISI rcvd at 9-APR-77 0037-PST Date: 9 APR 1977 0021-PST Sender: STEFFERUD at USC-ISI Subject: MSGGROUP# 489 $Add DDeutsch@BBNA, Mark.Faust@CMUA, Remove Wulf@ISIB $ From: MSGGROUP at USC-ISI To: [ISI]Mailing.List: Cc: Wulf at ISIB Message-ID: <[USC-ISI] 9-APR-77 00:21:14.STEFFERUD> For those who do not know of my convention, the $----$ brackets in the subject indicate that there is nothing else of interest in the text of the message. It is all contained in the subject. This message violates the convention in order to explain the convention, Stef ------- 10-MAY-78 22:13:46-PDT,14335;000000000001 Mail from OFFICE-1 rcvd at 11-APR-77 1306-PST Date: 11 APR 1977 1303-PST Sender: LEDUC at OFFICE-1 Subject: MSGGROUP# 490 Electronic Message systems for the U.S. postal service From: LEDUC at OFFICE-1 To: msggroup@ISI Message-ID: <[OFFICE-1]11-APR-77 13:02:32-PST.LEDUC> Stef, this is the summary of the presentation on Electronic Message Systems for the U.S. Postal Service that was given to Congress by Harold Belcher. A full description of the situation is given in the NTIS publication bearing the same name, and having the order no. :PB 2622892. the report was issued in 1976. Nicole PS: From this point on, the format includes page breaks each 66 lines, indicated by "---------" at the left margin. < MEADE, USPS.NLS;1, >, 25-APR-77 13:26 HEM ;;;; --------- REPORT: ELECTRONIC MESSAGE SYSTEMS FOR THE U.S. POSTAL SERVICE OBJECTIVES: 1. The goal of the study was to recommend a course of action to the U.S.P.S. with regard to the research, development and applications of communications-information processing technologies and electronic transmission and delivery systems to supplement or partly replace first class letter mail. 2. Propose U.S.P.S. strategies for Electronic Message Services. 3. Examine some of the effects that Electronic Message Services are likely to have on such matters as privacy, economics, and regulation of the telecommunications industry. 4. Reviewed future domestic and international telecommunications system and equipment concepts that could be used by the U.S.P.S. for Electronic Message Services in the 1985 to 1990 period. 1 --------- REPORT: ELECTRONIC MESSAGE SYSTEMS FOR THE U.S. POSTAL SERVICE STRATEGY: 1. Limit the role of the U.S.P.S. to "Business as usual" -- the physical handling of letters, circulars, periodicals, and parcels and sale of postage stamps. This strategy would leave the transfer of messages by electronic means to the communications and information processing industries. FINDINGS: The panel believes that this strategy -- LIMITING THE U.S.P.S. ROLE TO PHYSICAL HANDLING OF MAIL -- will lead to greater costs and poorer service -- conditions that are unacceptable to the nation. 2. Commit the U.S.P.S. to a future involving the transfer of messages by electronic means, in accordance with either or both of the following options: A. Develop ventures in cooperation with communications carriers, information processing services, and users, capitalizing on the existing strengths of the U.S.P.S. B. Provide new electronic services to improve the collection, transmission, distribution, and delivery systems, accepting the fact that this option may produce policy and competition issues in connection with existing and planned communication carrier offerings. FINDINGS: The application of electronic technology to the mail service is no panacea, but it does appear to offer the U.S.P.S. an opportunity to turn away from a course that shows little or no promise of improvement. 2 --------- REPORT: ELECTRONIC MESSAGE SYSTEMS FOR THE U.S. POSTAL SERVICE CONCLUSIONS: The panel concludes that: 1. Technological feasibility does not limit the evolutionary development of Electronic Message Services. Instead, the success or failure of such services depend upon combinations of technical, societal, economic and regulatory factors. Technologies that are likely to be used in providing an Electronic Message Service within the next 10 to 15 years are already available or in development. Nevertheless, modifications and developments of equipments and software are required to arrive at Electronic Message Systems that are reliable, versatile and cost-effective. 2. Because several large electronic message distribution systems are already in successful operation, any U.S.P.S. planning and implementation of Electronic Message Services may benefit from the experience gained in establishing and operating these existing systems. 3. It is not necessary and may not be desirable that the U.S.P.S. own and operate a complete message system. Subsystems could be either leased or owned, depending upon specific analyses in each case. Because of the dynamic and evolutionary nature of the network, the entire system will need to be managed by the U.S.P.S. to be effective. 4. Electronic Message Systems offer the potential of replacing as much as one-third of all of today's letter mail, which included nearly one-half of all first-class mail. Most of the electronic messages originated by private individuals will probably make up less than 10 percent of the total. Accordingly, the requirements for local letter carrier distribution will not decrease significantly in the near future. 5. Regulatory issues that arise as a result of the implementation of Generation I and II can probably be dealt with under existing procedures. Major difficulties are not expected. However, the implementation of Generation III will undoubtedly raise difficult and complex national policy issues concerning competition and regulation. A 3 --------- court test may be required to settle such issues - or the Congress may have to take up the matter. Security and privacy implications are inherent in all three generations and need to be carefully considered. 6. Several other countries, including Canada, the United Kingdom, West Germany and Japan, are proceeding in varying degrees along lines generally consistent with the conclusions of this panel that present mail processing can be significantly supplemented or replaced by Electronic Message Services. 7. Present and proposed international telecommunications systems are capable of accommodating all presently foreseen requirements for electronic message transmission. 4 --------- REPORT: ELECTRONIC MESSAGE SYSTEMS FOR THE U.S. POSTAL SERVICE RECOMMENDATIONS: The panel recommends that: 1. Among the alternatives to the U.S.P.S., Electronic Message Services offer an opportunity to reverse the present trend of rising costs, decreasing volume, and increasing deficits. Electronic Message Systems are neither a panacea nor a guaranteed solution to the present problems of the Postal Service but if a significant portion of today's first-class letter mail is replaced by such systems, the U.S.P.S. might find it possible to achieve economies. 2. U.S.P.S. top management needs to adopt a firm and continuing commitment to involvement in the electronic message field. The commitment that is called for is the same type of conviction in the eventual success of the project that was prevalent within the top management of the National Aeronautical and Space Administration (NASA) for the manned space program in the 1960's. 3. U.S.P.S. management needs to place increased emphasis on planning, research and development, systems engineering, and in-house capability. 4. Some demonstration field trials will be required to develop and test the new systems, as well as to determine the user acceptance of the services. It may be desirable for the U.S.P.S. to obtain the participation of business users or government agencies in such field trials. It is not certain that present mail patterns can yield good predictions of an Electronic Message System. The response of prospective users to a demonstration of Electronic Message Services is necessary to decide on wider applications. Not only can user reactions be ascertained in field trials, but the likelihood of developing a more acceptable public service would be enhanced. There are too many factors involved to determine the economic feasibility of the service a priori. It may be that the U.S.P.S. will not benefit economically until a Generation III system is operating, but Generations I and II need to be in place, at least in prototype, before Generation III is feasible. 5 --------- 5. At the earliest opportunity, the Congress would do well to address the adoption of a policy on Electronic Message Services that best serves the nation. The mail is a matter of national policy, affecting an essential service that has been in the hands of the Federal Government since its founding. The Congress, therefore, will have to consider the function of the U.S.P.S., especially its record in handling the technology of Generation I and II, and the situation with regard to the private communications carriers. The panel recognizes that this is a difficult issue -- one that certainly includes complicated social, economic, legal, and political questions. For most countries, electronic communications happens to be a government monopoly, like the Postal Service. In the U.S., by contrast, it has been a matter of national policy that all electronic communications, except those under military jurisdiction, have been left to private enterprise. Yet any system using the sophisticated Generation III concept would blur today's clear distinction between the mail and the service rendered by the electronic communications common carriers. 6 --------- COMMITTEE ON TELECOMMUNICATION U.S.P.S. SUPPORT PANEL MEMBERS * Louis T. Rader (Chairman) Professor of Electrical Engineering and Business Administration University of Virginia * Henri Busignies Chief Scientist Emeritus International Telephone & Telegraph Corporation * Charles H. Elmendorf, III Assistant Vice President- Engineering Department American Telephone & Telegraph Corp. * William L. Everitt Dean Emeritus of Engineering Univ. of Illinois at Urbana Guy Fougere Vice President, Corporate Staff Arthur D. Little, Inc. Henry Geller Communications Attorney Aspen Institute Program on Communications & Society James J. Johnson Executive Vice President (until May 1976) Western Union Telegraph Co. * Sidney Metzger Assistant Vice President and Chief Scientist Communications Satellite Corp. Alvin E. Nashman President, Systems Division Computer Sciences Corp. Charles W. Nicolson Vice President Engineering Services Potomac Electric Power Co. John S. Reed Executive Vice President First National City Bank 7 --------- Emmett J. Rice Senior Vice President The National Bank of Washington, D.C. Bernard Strassburg Communications Consultant Former Chief, Common Carrier Bu. Federal Communications Commission H. Mitchell Watson, Jr. Vice-President, Communications Technology & Manufacturing Systems Development Division IBM Corporation PANEL CONSULTANTS: Armig G. Kandoian Thomas A. Prugh * Member of the National Academy of Engineering --------- REPORT OF: U.S.P.S. SUPPORT PANEL COMMITTEE ON TELECOMMUNICATIONS ASSEMBLY OF ENGINEERING ON ELECTRONIC MESSAGE SYSTEMS FOR THE U.S. POSTAL SERVICE JANUARY 1977 --------- DEFINITION OF GENERATIONS I, II III, MAIL STREAMS GENERATION I Originator to Post Office to Sectional Center Facility by Hard Copy Transportation; Sectional Center Facility to Sectional Center Facility by Electronic Transmission Sectional Center Facility to Post Office to Recipient by Hard Copy Transportation GENERATION II Originator to Electronic Message System by Electronic Input; Electronic Message System to Appropriate USPS Installation by Electronic Transmission; Appropriate USPS Installation to Recipient by Hard Copy Transportation GENERATION III Originator to Electronic Message System to Recipient by Electronic Transmission --------- ------- 10-MAY-78 22:13:46-PDT,1426;000000000001 Mail from OFFICE-1 rcvd at 12-APR-77 1126-PST Date: 12 APR 1977 1120-PST From: WALSH at OFFICE-1 Subject: MSGGROUP# 491 Introduction of Rich Zellich (WALSH AT OFFICE-1) To: MSGGROUP at ISI cc: WALSH Greetings - I am a chief programer/team leader in the Logistics Management Information Division of the Automated Logistics Management Systems Activity (ALMSA), a central systems design activity (CSDA) under Hq, DARCOM> ALMSA is located in St. Louis, Mo and has been in existence for 8 or 9 years. In the last 3-4 years we (my division in particular) have finally goten into interactive systems, and now we are also entering the networking arena. My roles in message systems are: 1. As a user and critic, with a strong interest in future network systems for use by the general public. 2. As an implementer, in the sense that I am part of designing/ writing multiple interactive applications that will be marketed to a user comunity (the 56 DARCOM project management offices and, eventually, other organizations) as a package. This marketing package will also contain message/filing/text-editing/etc. applications that will have been developed by other groups, but that must appear to the users as consistent with the rest of the package. I can be reached through the WALSH AT OFFICE-1 mailbox or, for NLS users, with IDENT=RICH. Regards to all, Rich Zellich ------- 10-MAY-78 22:13:47-PDT,4579;000000000001 Mail from OFFICE-1 rcvd at 13-APR-77 1029-PST Date: 13 APR 1977 1025-PST From: WALSH at OFFICE-1 Subject: MSGGROUP# 492 Intuitive commands & suggestions on FROM/TO/CC/FCC/SHOW To: MSGGROUP at ISI cc: WALSH, STEFFERUD at ISI 1. INTRODUCTION: I am new to MSGGROUP and have had time to read only a small percentage of the back files. Not currently having the time to read the remainder, I therefore apologize if some of the following comments seem to be beating dead horses or reinventing wheels. 2. NATURAL LANGUAGE/INTUITIVE COMMANDS: I am a firm believer in human-engineered/natural/intuitive interactive systems, but I disagree with some of the comments in MSGGROUP concerning "letters in envelopes", file cabinets, file drawers, file folders, etc. As I see it, the idea is to augment the office environment, not simulate it. The new tools we are offering provide users with new and better ways of doing things. Filing cabinets, drawers, folders do not possess many of the capabilities of the message/filing systems, so why try to present these systems in terms of a traditional office? "Natural" or "intuitive" commands can relate to a new way of doing things as well as to an old way. Certainly, some users will have to be educated in computer file/storage concepts, but that shouldn't be very difficult. We do not want to limit or over-simplify our systems just for non- computer users, because more and more new users will already be used to working with computer systems and will understand their background concepts and capabilities. 3. SHOW: This is a good general-purpose command for many situations, and is at least as intuitive as PRINT, TYPE, DISPLAY, WRITE, etc. It can be used to "show" status, headers, messages, options, directories, or just about anything else and is equally applicable to CRT's and to typewriters. Extensively used, it could become almost as useful as HELP when learning a new system. 4. PROMPTING(message sending): I would like to see the prompts for SUBJECT, TO, CC, etc. come after message-text creation rather than before. This is not so important when sending a pre-built message/file or when you have the option of re-specifying these fields at any time, but when using a simple-minded system and composing on the fly, too often the final message content causes after-thoughts about the title/subject or who should be addressed for action or copies. 5. TO/CC: How about a "?" option for a more sophisticated system, to search the current directory for group addresses, with the possible sub-option of asking for a full membership display of a tentatively- selected group before either accepting or rejecting. 6. ATTN: This subfield is needed as part of any message system, no matter how primitive, but I would like to see the generated "ATTN:" eliminated. This would let the sender put any text string following the address. Examples: WALSH@OFFICE-1(ACTION) WALSH@OFFICE-1(RICH Z.) WALSH@OFFICE-1(ATTN: R. ZELLICH & W. MARTIN). 7. FROM/SENDER: The elegant solution here would be to have a single "SENDER" header-field, but with the addition of an "ATTN subfield". This could be prompted as FROM:, and would generate the mailbox address followed by the subfield containing whatever "from" text string that had been entered. To prompt for FROM: or not should of course be user-selectable via the profile. Examples of use: FROM: FROM: RICH Z. FROM: R. ZELLICH FOR W. WALSH The receiver then would see something like: SENDER: WALSH@OFFICE-1 or SENDER: WALSH@OFFICE-1(RICH Z.) or SENDER: WALSH@OFFICE-1(R. ZELLICH FOR W. WALSH) For those requiring such fields as "authorized-by", "requested-by", "released-by", "for", etc., rather than build unique message systems, I recommend such fields appear as text and that they be handled by user FORMS subsystems when possible. Certainly an open-ended "include anything" header format can be defined, but why add the additional complexity to some sending and all receiving systems when it's not really necessary? 8. FILE/FCC: This would be an excellent place to allow a program- name as well as a file-name, as a means of implementing user-written automatic filing systems. Rich Zellich ------- 10-MAY-78 22:13:47-PDT,441;000000000001 Mail from USC-ISI rcvd at 13-APR-77 2338-PST Date: 13 APR 1977 2328-PST Sender: STEFFERUD at USC-ISI Subject: MSGGROUP# 493 $Add Brotz@PARC-MAXC, RGH@SU-AI(Richard Henninger)$ From: MSGGROUP at USC-ISI To: [ISI]Mailing.List: Cc: Brotz at PARC-MAXC, RGH at SU-AI(Attn: Richard.Henninger) Message-ID: <[USC-ISI]13-APR-77 23:28:26.STEFFERUD> Stef PS: Subject: $----$ means all content is in subject field. ------- 10-MAY-78 22:13:47-PDT,931;000000000001 Mail from USC-ISI rcvd at 14-APR-77 0029-PST Date: 14 APR 1977 0029-PST Sender: STEFFERUD at USC-ISI Subject: MSGGROUP# 494 [BROTZ at PARC-MAXC: Introduction] From: MSGGROUP at USC-ISI To: MsgGroup Message-ID: <[USC-ISI]14-APR-77 00:29:34.STEFFERUD> Begin forwarded message -------------------- Date: 12 APR 1977 1549-PST From: BROTZ at PARC-MAXC To: STEFFERUD at USC-ISI Cc: brotz Subject: Introduction and Dist. list Mail from PARC-MAXC rcvd at 12-APR-77 1550-PST I am working in Communications Systems at Xerox Systems Development Division in Palo Alto. One of the areas I will be working in is Electronic Mail. I would appreciate it very much if you could put me on the Mailing.List so that I could receive future correspondence on Electronic Mail. Thank you. --Doug Brotz at PARC-MAXC ------- -------------------- End forwarded message ------- 10-MAY-78 22:13:47-PDT,1555;000000000001 Mail from OFFICE-1 rcvd at 14-APR-77 1014-PST Date: 14 APR 1977 1013-PST From: PANKO at OFFICE-1 Subject: MSGGROUP# 495 Graphnet To: [ISI]Mailing.List: cc: rulifson at PARC-MAXC, sutherland at PARC-MAXC, cc: taylor at PARC-MAXC, bedford I have had a couple of inquiries about Graphnet, so I thought I'd give some information onGraphnet's mail service, FAXGRAM. For further information, call (800) 631-1581. Graphnet, in theory, is a facsimile transmission network that takes input from any fax terminal, compresses it, and sends it to any other fax terminal. But the costs of this service are absurdly high. A 12-page letter sent from Boston to Chicago, according to Howard Anderson of the Ynakee Group, is $144. What Graphnet really is is a direct competitor of mailgram and telegram. It accepts information from teletypewriter services like Telex and TWX (cheap), telephone calls (more expensive) and facsimile (very expensive). The resultant FAXGRAM is either telephoned or messengered to receivers. A $150-word FAXGRAM costs $5.03 if delivered by phone if the input medium is Telex or TWX. If the message is originated by telephone or fax and delivered by telephone, the charge is $13.83. If originated by Telex or TWX and delivered by messenger, the cost is $7.95. If it is originated by Fax or telephone and messngered, the cost is $16.75. That is a 150-word message, not $150-word message. This is cheaper than full-rate telegram. But who sends those any more? Ra3y Panko ------- 10-MAY-78 22:13:47-PDT,1755;000000000001 Mail from MIT-AI rcvd at 14-APR-77 1259-PST Date: 14 APR 1977 1557-EST From: RMS at MIT-AI (Richard M. Stallman ) Subject: MSGGROUP# 496 CONFUSION RE #492 FROM/SENDER To: STEFFERUD at USC-ISI I just got some mail to MSGGROUP which it is not at all clear how I should reply to. The "Sender" was you, the "From" was someone named Walsh, and then in parentheses there was somebody else's name. That name seemed to be the name of the real sender but there was no suggestion of how to reply to him. What I wanted to reply to him about was "From:" and "Attn:". It seems that the most important thing that a message must say is the name of the person to whom replies should go. I am not sure any more what "From" and "Sender" are supposed to mean, but in this case at least neither one seemed to tell me how to reply. In AIMLMC local mail we provide two names: the name of the person to reply to in mail, and the name that he was logged in as if it wasn't the same. If you send a message from someone else's terminal, you just tell the mail composer who you really are. The message comes out saying it is from you, with the name of the person whose terminal you used in parentheses. Thus, reply mail will go to you, while if the recipient wants to communicate with you instantly he can tell what name to link to. I have often wished that people who sent mail to me from other systems had used such a feature, instead of putting their (useless) mundane names at the bottom of the message. I claim that this represents the true structure of the various types of author of the message: the person(s) to reply to is the most important of them all, and the others should be present only if they differ from him. ------- 10-MAY-78 22:13:47-PDT,1495;000000000001 Mail from USC-ISI rcvd at 14-APR-77 1552-PST Date: 14 APR 1977 1542-PST Sender: STEFFERUD at USC-ISI Subject: MSGRROUP# 497 Re: CONFUSION RE $492 FROM/SENDER From: STEFFERUD at USC-ISI To: RMS at MIT-AI Cc: STEFFERUD, Walsh at OFFICE-1(Attn: Zellich) Message-ID: <[USC-ISI]14-APR-77 15:42:34.STEFFERUD> In-Reply-To: Your message of APRIL 14, 1977 Hi Richard, My sympathies to you. I guess I did not worry about it in the terms you describe. What I did was take a message that Rich Zellich sent to MsgGroup at ISI for redistribution, and I EXPLODED it with HERMES, which is to say, put it back into the compose buffers as though I had composed it. That put Walsh at Office-1 in the from field. I then edited the from field to show that Rich Zellich is the person in Walsh@O-1 who really sent it. Unfortunately, all those meanings are assumed in my head, and not in all other heads, as you make so very clear. Now then, if you wish, I will redistribute your reply to all MsgGroup, forwarding it so as to avoid such confusions again, or I will forward it to Rich Zellich via Walsh@O-1. Please indicate your choice. I am inclined to ship the whole thing to all MsgGroup cause I think your points are valid, and the situation is no doubt pervasive for the other folks who got that message. I am copying Walsh@O-1 on this in any case, and should forward your message to Rich since that is where you really wanted it to go. Stef ------- 10-MAY-78 22:13:47-PDT,1086;000000000001 Mail from MIT-AI rcvd at 14-APR-77 2057-PST Date: 14 APR 1977 2356-EST From: RMS at MIT-AI (Richard M. Stallman ) Subject: MSGGROUP# 498 MORE Re: CONFUSION RE: $492 FROM/SENDER To: Stefferud at USC-ISI Please forward my message to all of MSGGROUP (and this one). What I want to emphasize is that Rich Zellich should have had a feature he could use to cause his message to say, in every way, that the message was from him, despite the fact that he was logged in as "Walsh". Rather than "It's from Walsh. No, forget that, it's from Zellich", we should have seen "It's from Zellich (but sent from Walsh's terminal)", so that program which says "In 2 words or less, who is it from?" would get "Zellich" and not "Walsh" as the answer. If you (Stef) had used a similar feature in forwarding the message, that might also have been a win. Of course, the existence of such a feature would make some paranoid, security-minded types shudder, but in most places the "danger" is not as big as the inconvenience and confusion that this feature would eliminate. ------- 10-MAY-78 22:13:48-PDT,1077;000000000001 Mail from SUMEX-AIM rcvd at 15-APR-77 1148-PST Date: 15 APR 1977 1123-PST From: Kahler at SUMEX-AIM Subject: MSGGROUP# 499 Re: Intuitive commands and suggestions To: [ISI]Mailing.List: With reference to item 4 of Rich Zellich's comments: POST (the bulletin poster for the bulletin boards at SUMEX and ECL) has a nice feature. Control-UP (^^, '36) lets the user start POST over with what he has typed in still intact. He may review his response to each prompt and leave it as is or edit it as he wishes. (The program also offers a "reserve buffer" which is used to hold the latter portion of any of these segments while the user edits in the middle. It is not much to read about but is quite something to see, if I do say so myself.) Another control character types out all segments of the message from the beginning, as ^S types out the current segment. The user is free, as in SNDMSG, to type in the message first and TO/CC later. We don't have HERMES here at SUMEX-AIM. Someday I will have to try it out. Rich ------- 10-MAY-78 22:13:48-PDT,2257;000000000001 Mail from OFFICE-1 rcvd at 15-APR-77 1232-PST Date: 15 APR 1977 1215-PST Sender: WALSH at OFFICE-1 Subject: MSGGROUP# 500 FROM/SENDER (IN REPLY TO STALLMAN COMMENTS) From: WALSH at OFFICE-1 To: MSGGROUP at USC-ISI Cc: STEFFERUD at USC-ISI, RMS at MIT-AI, WALSH(Attn: RICH) Message-ID: <[OFFICE-1]15-APR-77 12:15:29.WALSH> THE CONFUSION OVER WHO SENT YOU THE ORIGINAL MESSAGE IS WHAT I WAS TRYING TO AVOID WITH THE PROPOSAL FOR A SINGLE "FROM" FIELD. MULTIPLE FROM, SENDER, AUTHOR, ETC. FIELDS SERVE MORE TO CONFUSE THAN TO CLARIFY. REPLY-TO IS CLEAR, BUT IS UNNECESSARY IF A FROM-AND-SUBFIELD CONVENTION IS ADOPTED. NOTE THAT IF I HAD A MAILBOX OF MY OWN AND WAS MERELY BORROWING BILL WALSH'S, I WOULD HAVE [LIKE TO HAVE] TYPED "FROM: WALSH@OFFICE-1(ZELLICH@OFFICE-1)" INSTEAD OF "FROM: WALSH@OFFICE-1(RICH ZELLICH)". THIS ALWAYS LEAVES IT CLEAR AS TO WHOM THE REPLY SHOULD BE SENT. IF I AM CORRECTLY INTERPRETING THE WAY MIT-AI'S MAILER WORKS, IT USES THE OPPOSITE CONVENTION: REAL NAME/MAILBOX FIRST, FOLLOWED BY MAILBOX ACTUALLY SENT FROM IN THE PARENTHETICAL SUBFIELD. THE REASON I DON'T PREFER THIS VARIATION IS THAT WHEN YOU USE THE REPLY/ANSWER FACILITY, THE MAILER CAN NO LONGER PICK UP THE ORIGINAL FROM FIELD AND SUBFIELD AND USE IT FOR A "TO" ADDRESS IN THE REPLY (UNLESS, OF COURSE, WE ALSO SWITCH THE "TO" FIELD AND ATTN SUBFIELD). IN THE ODD CASE WHERE, FOR SOME REASON, THE NAMES/MAILBOXES IN THE FROM FIELD AND SUBFIELD DON'T TRULY INDICATE WHO SHOULD RECEIVE THE ANSWER, "REPLY TO: XXXX@YYYY" CAN ALWAYS BE ENTERED IN THE TEXT (OR PROMPTED FOR BY A "FORMS SUBSYSTEM" WHICH WOULD STILL INCLUDE THE ENTRY AS PART OF THE TEXT). ACTUALLY, THERE'S NO REASON WHY THE FROM FIELD CAN'T LOOK LIKE "FROM: WALSH@OFFICE-1(ZELLICH@OFFICE-1(REPLY TO RHILL@BBNB))". INCIDENTALLY, STEF'S "ATTN SUBFIELD" CC ENTRY HIGHLIGHTS WHAT I WAS TALKING ABOUT WHEN I SUGGESTED ELIMINATING THE MAILER-GENERATED "ATTN:" STRING IN THE SUBFIELD. STEF TYPED "RMS AT MIT-AI (JUST-SO-YOU-KNOW-WHAT-I-DID-OK)" IN THE CC FIELD, AND IT CAME OUT "RMS AT MIT-AI (ATTN: JUST-SO-YOU-KNOW-WHAT-I-DID-OK)". WALSH@OFFICE-1(RICH ZELLICH) -------